3 perlhack - How to hack at the Perl internals
7 This document attempts to explain how Perl development takes place,
8 and ends with some suggestions for people wanting to become bona fide
11 The perl5-porters mailing list is where the Perl standard distribution
12 is maintained and developed. The list can get anywhere from 10 to 150
13 messages a day, depending on the heatedness of the debate. Most days
14 there are two or three patches, extensions, features, or bugs being
17 A searchable archive of the list is at either:
19 http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters/
23 http://archive.develooper.com/perl5-porters@perl.org/
25 List subscribers (the porters themselves) come in several flavours.
26 Some are quiet curious lurkers, who rarely pitch in and instead watch
27 the ongoing development to ensure they're forewarned of new changes or
28 features in Perl. Some are representatives of vendors, who are there
29 to make sure that Perl continues to compile and work on their
30 platforms. Some patch any reported bug that they know how to fix,
31 some are actively patching their pet area (threads, Win32, the regexp
32 engine), while others seem to do nothing but complain. In other
33 words, it's your usual mix of technical people.
35 Over this group of porters presides Larry Wall. He has the final word
36 in what does and does not change in the Perl language. Various
37 releases of Perl are shepherded by a "pumpking", a porter
38 responsible for gathering patches, deciding on a patch-by-patch,
39 feature-by-feature basis what will and will not go into the release.
40 For instance, Gurusamy Sarathy was the pumpking for the 5.6 release of
41 Perl, and Jarkko Hietaniemi was the pumpking for the 5.8 release, and
42 Rafael Garcia-Suarez holds the pumpking crown for the 5.10 release.
44 In addition, various people are pumpkings for different things. For
45 instance, Andy Dougherty and Jarkko Hietaniemi did a grand job as the
46 I<Configure> pumpkin up till the 5.8 release. For the 5.10 release
47 H.Merijn Brand took over.
49 Larry sees Perl development along the lines of the US government:
50 there's the Legislature (the porters), the Executive branch (the
51 pumpkings), and the Supreme Court (Larry). The legislature can
52 discuss and submit patches to the executive branch all they like, but
53 the executive branch is free to veto them. Rarely, the Supreme Court
54 will side with the executive branch over the legislature, or the
55 legislature over the executive branch. Mostly, however, the
56 legislature and the executive branch are supposed to get along and
57 work out their differences without impeachment or court cases.
59 You might sometimes see reference to Rule 1 and Rule 2. Larry's power
60 as Supreme Court is expressed in The Rules:
66 Larry is always by definition right about how Perl should behave.
67 This means he has final veto power on the core functionality.
71 Larry is allowed to change his mind about any matter at a later date,
72 regardless of whether he previously invoked Rule 1.
76 Got that? Larry is always right, even when he was wrong. It's rare
77 to see either Rule exercised, but they are often alluded to.
79 New features and extensions to the language are contentious, because
80 the criteria used by the pumpkings, Larry, and other porters to decide
81 which features should be implemented and incorporated are not codified
82 in a few small design goals as with some other languages. Instead,
83 the heuristics are flexible and often difficult to fathom. Here is
84 one person's list, roughly in decreasing order of importance, of
85 heuristics that new features have to be weighed against:
89 =item Does concept match the general goals of Perl?
91 These haven't been written anywhere in stone, but one approximation
94 1. Keep it fast, simple, and useful.
95 2. Keep features/concepts as orthogonal as possible.
96 3. No arbitrary limits (platforms, data sizes, cultures).
97 4. Keep it open and exciting to use/patch/advocate Perl everywhere.
98 5. Either assimilate new technologies, or build bridges to them.
100 =item Where is the implementation?
102 All the talk in the world is useless without an implementation. In
103 almost every case, the person or people who argue for a new feature
104 will be expected to be the ones who implement it. Porters capable
105 of coding new features have their own agendas, and are not available
106 to implement your (possibly good) idea.
108 =item Backwards compatibility
110 It's a cardinal sin to break existing Perl programs. New warnings are
111 contentious--some say that a program that emits warnings is not
112 broken, while others say it is. Adding keywords has the potential to
113 break programs, changing the meaning of existing token sequences or
114 functions might break programs.
116 =item Could it be a module instead?
118 Perl 5 has extension mechanisms, modules and XS, specifically to avoid
119 the need to keep changing the Perl interpreter. You can write modules
120 that export functions, you can give those functions prototypes so they
121 can be called like built-in functions, you can even write XS code to
122 mess with the runtime data structures of the Perl interpreter if you
123 want to implement really complicated things. If it can be done in a
124 module instead of in the core, it's highly unlikely to be added.
126 =item Is the feature generic enough?
128 Is this something that only the submitter wants added to the language,
129 or would it be broadly useful? Sometimes, instead of adding a feature
130 with a tight focus, the porters might decide to wait until someone
131 implements the more generalized feature. For instance, instead of
132 implementing a "delayed evaluation" feature, the porters are waiting
133 for a macro system that would permit delayed evaluation and much more.
135 =item Does it potentially introduce new bugs?
137 Radical rewrites of large chunks of the Perl interpreter have the
138 potential to introduce new bugs. The smaller and more localized the
141 =item Does it preclude other desirable features?
143 A patch is likely to be rejected if it closes off future avenues of
144 development. For instance, a patch that placed a true and final
145 interpretation on prototypes is likely to be rejected because there
146 are still options for the future of prototypes that haven't been
149 =item Is the implementation robust?
151 Good patches (tight code, complete, correct) stand more chance of
152 going in. Sloppy or incorrect patches might be placed on the back
153 burner until the pumpking has time to fix, or might be discarded
154 altogether without further notice.
156 =item Is the implementation generic enough to be portable?
158 The worst patches make use of a system-specific features. It's highly
159 unlikely that non-portable additions to the Perl language will be
162 =item Is the implementation tested?
164 Patches which change behaviour (fixing bugs or introducing new features)
165 must include regression tests to verify that everything works as expected.
166 Without tests provided by the original author, how can anyone else changing
167 perl in the future be sure that they haven't unwittingly broken the behaviour
168 the patch implements? And without tests, how can the patch's author be
169 confident that his/her hard work put into the patch won't be accidentally
170 thrown away by someone in the future?
172 =item Is there enough documentation?
174 Patches without documentation are probably ill-thought out or
175 incomplete. Nothing can be added without documentation, so submitting
176 a patch for the appropriate manpages as well as the source code is
179 =item Is there another way to do it?
181 Larry said "Although the Perl Slogan is I<There's More Than One Way
182 to Do It>, I hesitate to make 10 ways to do something". This is a
183 tricky heuristic to navigate, though--one man's essential addition is
184 another man's pointless cruft.
186 =item Does it create too much work?
188 Work for the pumpking, work for Perl programmers, work for module
189 authors, ... Perl is supposed to be easy.
191 =item Patches speak louder than words
193 Working code is always preferred to pie-in-the-sky ideas. A patch to
194 add a feature stands a much higher chance of making it to the language
195 than does a random feature request, no matter how fervently argued the
196 request might be. This ties into "Will it be useful?", as the fact
197 that someone took the time to make the patch demonstrates a strong
198 desire for the feature.
202 If you're on the list, you might hear the word "core" bandied
203 around. It refers to the standard distribution. "Hacking on the
204 core" means you're changing the C source code to the Perl
205 interpreter. "A core module" is one that ships with Perl.
207 =head2 Keeping in sync
209 The source code to the Perl interpreter, in its different versions, is
210 kept in a repository managed by a revision control system ( which is
211 currently the Perforce program, see http://perforce.com/ ). The
212 pumpkings and a few others have access to the repository to check in
213 changes. Periodically the pumpking for the development version of Perl
214 will release a new version, so the rest of the porters can see what's
215 changed. The current state of the main trunk of repository, and patches
216 that describe the individual changes that have happened since the last
217 public release are available at this location:
219 http://public.activestate.com/pub/apc/
220 ftp://public.activestate.com/pub/apc/
222 If you're looking for a particular change, or a change that affected
223 a particular set of files, you may find the B<Perl Repository Browser>
226 http://public.activestate.com/cgi-bin/perlbrowse
228 You may also want to subscribe to the perl5-changes mailing list to
229 receive a copy of each patch that gets submitted to the maintenance
230 and development "branches" of the perl repository. See
231 http://lists.perl.org/ for subscription information.
233 If you are a member of the perl5-porters mailing list, it is a good
234 thing to keep in touch with the most recent changes. If not only to
235 verify if what you would have posted as a bug report isn't already
236 solved in the most recent available perl development branch, also
237 known as perl-current, bleading edge perl, bleedperl or bleadperl.
239 Needless to say, the source code in perl-current is usually in a perpetual
240 state of evolution. You should expect it to be very buggy. Do B<not> use
241 it for any purpose other than testing and development.
243 Keeping in sync with the most recent branch can be done in several ways,
244 but the most convenient and reliable way is using B<rsync>, available at
245 ftp://rsync.samba.org/pub/rsync/ . (You can also get the most recent
248 If you choose to keep in sync using rsync, there are two approaches
253 =item rsync'ing the source tree
255 Presuming you are in the directory where your perl source resides
256 and you have rsync installed and available, you can "upgrade" to
259 # rsync -avz rsync://public.activestate.com/perl-current/ .
261 This takes care of updating every single item in the source tree to
262 the latest applied patch level, creating files that are new (to your
263 distribution) and setting date/time stamps of existing files to
264 reflect the bleadperl status.
266 Note that this will not delete any files that were in '.' before
267 the rsync. Once you are sure that the rsync is running correctly,
268 run it with the --delete and the --dry-run options like this:
270 # rsync -avz --delete --dry-run rsync://public.activestate.com/perl-current/ .
272 This will I<simulate> an rsync run that also deletes files not
273 present in the bleadperl master copy. Observe the results from
274 this run closely. If you are sure that the actual run would delete
275 no files precious to you, you could remove the '--dry-run' option.
277 You can than check what patch was the latest that was applied by
278 looking in the file B<.patch>, which will show the number of the
281 If you have more than one machine to keep in sync, and not all of
282 them have access to the WAN (so you are not able to rsync all the
283 source trees to the real source), there are some ways to get around
288 =item Using rsync over the LAN
290 Set up a local rsync server which makes the rsynced source tree
291 available to the LAN and sync the other machines against this
294 From http://rsync.samba.org/README.html :
296 "Rsync uses rsh or ssh for communication. It does not need to be
297 setuid and requires no special privileges for installation. It
298 does not require an inetd entry or a daemon. You must, however,
299 have a working rsh or ssh system. Using ssh is recommended for
300 its security features."
302 =item Using pushing over the NFS
304 Having the other systems mounted over the NFS, you can take an
305 active pushing approach by checking the just updated tree against
306 the other not-yet synced trees. An example would be
315 $1 => [ (stat $1)[2, 7, 9] ]; # mode, size, mtime
318 my %remote = map { $_ => "/$_/pro/3gl/CPAN/perl-5.7.1" } qw(host1 host2);
320 foreach my $host (keys %remote) {
321 unless (-d $remote{$host}) {
322 print STDERR "Cannot Xsync for host $host\n";
325 foreach my $file (keys %MF) {
326 my $rfile = "$remote{$host}/$file";
327 my ($mode, $size, $mtime) = (stat $rfile)[2, 7, 9];
328 defined $size or ($mode, $size, $mtime) = (0, 0, 0);
329 $size == $MF{$file}[1] && $mtime == $MF{$file}[2] and next;
330 printf "%4s %-34s %8d %9d %8d %9d\n",
331 $host, $file, $MF{$file}[1], $MF{$file}[2], $size, $mtime;
333 copy ($file, $rfile);
334 utime time, $MF{$file}[2], $rfile;
335 chmod $MF{$file}[0], $rfile;
339 though this is not perfect. It could be improved with checking
340 file checksums before updating. Not all NFS systems support
341 reliable utime support (when used over the NFS).
345 =item rsync'ing the patches
347 The source tree is maintained by the pumpking who applies patches to
348 the files in the tree. These patches are either created by the
349 pumpking himself using C<diff -c> after updating the file manually or
350 by applying patches sent in by posters on the perl5-porters list.
351 These patches are also saved and rsync'able, so you can apply them
352 yourself to the source files.
354 Presuming you are in a directory where your patches reside, you can
355 get them in sync with
357 # rsync -avz rsync://public.activestate.com/perl-current-diffs/ .
359 This makes sure the latest available patch is downloaded to your
362 It's then up to you to apply these patches, using something like
364 # last="`cat ../perl-current/.patch`.gz"
365 # rsync -avz rsync://public.activestate.com/perl-current-diffs/ .
366 # find . -name '*.gz' -newer $last -exec gzcat {} \; >blead.patch
368 # patch -p1 -N <../perl-current-diffs/blead.patch
370 or, since this is only a hint towards how it works, use CPAN-patchaperl
371 from Andreas König to have better control over the patching process.
375 =head2 Why rsync the source tree
379 =item It's easier to rsync the source tree
381 Since you don't have to apply the patches yourself, you are sure all
382 files in the source tree are in the right state.
384 =item It's more reliable
386 While both the rsync-able source and patch areas are automatically
387 updated every few minutes, keep in mind that applying patches may
388 sometimes mean careful hand-holding, especially if your version of
389 the C<patch> program does not understand how to deal with new files,
390 files with 8-bit characters, or files without trailing newlines.
394 =head2 Why rsync the patches
398 =item It's easier to rsync the patches
400 If you have more than one machine that you want to keep in track with
401 bleadperl, it's easier to rsync the patches only once and then apply
402 them to all the source trees on the different machines.
404 In case you try to keep in pace on 5 different machines, for which
405 only one of them has access to the WAN, rsync'ing all the source
406 trees should than be done 5 times over the NFS. Having
407 rsync'ed the patches only once, I can apply them to all the source
408 trees automatically. Need you say more ;-)
410 =item It's a good reference
412 If you do not only like to have the most recent development branch,
413 but also like to B<fix> bugs, or extend features, you want to dive
414 into the sources. If you are a seasoned perl core diver, you don't
415 need no manuals, tips, roadmaps, perlguts.pod or other aids to find
416 your way around. But if you are a starter, the patches may help you
417 in finding where you should start and how to change the bits that
420 The file B<Changes> is updated on occasions the pumpking sees as his
421 own little sync points. On those occasions, he releases a tar-ball of
422 the current source tree (i.e. perl@7582.tar.gz), which will be an
423 excellent point to start with when choosing to use the 'rsync the
424 patches' scheme. Starting with perl@7582, which means a set of source
425 files on which the latest applied patch is number 7582, you apply all
426 succeeding patches available from then on (7583, 7584, ...).
428 You can use the patches later as a kind of search archive.
432 =item Finding a start point
434 If you want to fix/change the behaviour of function/feature Foo, just
435 scan the patches for patches that mention Foo either in the subject,
436 the comments, or the body of the fix. A good chance the patch shows
437 you the files that are affected by that patch which are very likely
438 to be the starting point of your journey into the guts of perl.
440 =item Finding how to fix a bug
442 If you've found I<where> the function/feature Foo misbehaves, but you
443 don't know how to fix it (but you do know the change you want to
444 make), you can, again, peruse the patches for similar changes and
445 look how others apply the fix.
447 =item Finding the source of misbehaviour
449 When you keep in sync with bleadperl, the pumpking would love to
450 I<see> that the community efforts really work. So after each of his
451 sync points, you are to 'make test' to check if everything is still
452 in working order. If it is, you do 'make ok', which will send an OK
453 report to I<perlbug@perl.org>. (If you do not have access to a mailer
454 from the system you just finished successfully 'make test', you can
455 do 'make okfile', which creates the file C<perl.ok>, which you can
456 than take to your favourite mailer and mail yourself).
458 But of course, as always, things will not always lead to a success
459 path, and one or more test do not pass the 'make test'. Before
460 sending in a bug report (using 'make nok' or 'make nokfile'), check
461 the mailing list if someone else has reported the bug already and if
462 so, confirm it by replying to that message. If not, you might want to
463 trace the source of that misbehaviour B<before> sending in the bug,
464 which will help all the other porters in finding the solution.
466 Here the saved patches come in very handy. You can check the list of
467 patches to see which patch changed what file and what change caused
468 the misbehaviour. If you note that in the bug report, it saves the
469 one trying to solve it, looking for that point.
473 If searching the patches is too bothersome, you might consider using
474 perl's bugtron to find more information about discussions and
475 ramblings on posted bugs.
477 If you want to get the best of both worlds, rsync both the source
478 tree for convenience, reliability and ease and rsync the patches
483 =head2 Working with the source
485 Because you cannot use the Perforce client, you cannot easily generate
486 diffs against the repository, nor will merges occur when you update
487 via rsync. If you edit a file locally and then rsync against the
488 latest source, changes made in the remote copy will I<overwrite> your
491 The best way to deal with this is to maintain a tree of symlinks to
492 the rsync'd source. Then, when you want to edit a file, you remove
493 the symlink, copy the real file into the other tree, and edit it. You
494 can then diff your edited file against the original to generate a
495 patch, and you can safely update the original tree.
497 Perl's F<Configure> script can generate this tree of symlinks for you.
498 The following example assumes that you have used rsync to pull a copy
499 of the Perl source into the F<perl-rsync> directory. In the directory
500 above that one, you can execute the following commands:
504 ../perl-rsync/Configure -Dmksymlinks -Dusedevel -D"optimize=-g"
506 This will start the Perl configuration process. After a few prompts,
507 you should see something like this:
509 Symbolic links are supported.
511 Checking how to test for symbolic links...
512 Your builtin 'test -h' may be broken.
513 Trying external '/usr/bin/test -h'.
514 You can test for symbolic links with '/usr/bin/test -h'.
516 Creating the symbolic links...
517 (First creating the subdirectories...)
518 (Then creating the symlinks...)
520 The specifics may vary based on your operating system, of course.
521 After you see this, you can abort the F<Configure> script, and you
522 will see that the directory you are in has a tree of symlinks to the
523 F<perl-rsync> directories and files.
525 If you plan to do a lot of work with the Perl source, here are some
526 Bourne shell script functions that can make your life easier:
539 if [ -L $1.orig ]; then
545 Replace "vi" with your favorite flavor of editor.
547 Here is another function which will quickly generate a patch for the
548 files which have been edited in your symlink tree:
552 for f in `find . -name '*.orig' | sed s,^\./,,`
554 case `echo $f | sed 's,.orig$,,;s,.*\.,,'` in
556 pod) diffopts='-F^=' ;;
559 diff -du $diffopts $f `echo $f | sed 's,.orig$,,'`
563 This function produces patches which include enough context to make
564 your changes obvious. This makes it easier for the Perl pumpking(s)
565 to review them when you send them to the perl5-porters list, and that
566 means they're more likely to get applied.
568 This function assumed a GNU diff, and may require some tweaking for
571 =head2 Perlbug administration
573 There is a single remote administrative interface for modifying bug status,
574 category, open issues etc. using the B<RT> bugtracker system, maintained
575 by Robert Spier. Become an administrator, and close any bugs you can get
576 your sticky mitts on:
578 http://bugs.perl.org/
580 To email the bug system administrators:
582 "perlbug-admin" <perlbug-admin@perl.org>
584 =head2 Submitting patches
586 Always submit patches to I<perl5-porters@perl.org>. If you're
587 patching a core module and there's an author listed, send the author a
588 copy (see L<Patching a core module>). This lets other porters review
589 your patch, which catches a surprising number of errors in patches.
590 Either use the diff program (available in source code form from
591 ftp://ftp.gnu.org/pub/gnu/ , or use Johan Vromans' I<makepatch>
592 (available from I<CPAN/authors/id/JV/>). Unified diffs are preferred,
593 but context diffs are accepted. Do not send RCS-style diffs or diffs
594 without context lines. More information is given in the
595 I<Porting/patching.pod> file in the Perl source distribution. Please
596 patch against the latest B<development> version. (e.g., even if you're
597 fixing a bug in the 5.8 track, patch against the latest B<development>
598 version rsynced from rsync://public.activestate.com/perl-current/ )
600 If changes are accepted, they are applied to the development branch. Then
601 the 5.8 pumpking decides which of those patches is to be backported to the
602 maint branch. Only patches that survive the heat of the development
603 branch get applied to maintenance versions.
605 Your patch should update the documentation and test suite. See
606 L<Writing a test>. If you have added or removed files in the distribution,
607 edit the MANIFEST file accordingly, sort the MANIFEST file using
608 C<make manisort>, and include those changes as part of your patch.
610 Patching documentation also follows the same order: if accepted, a patch
611 is first applied to B<development>, and if relevant then it's backported
612 to B<maintenance>. (With an exception for some patches that document
613 behaviour that only appears in the maintenance branch, but which has
614 changed in the development version.)
616 To report a bug in Perl, use the program I<perlbug> which comes with
617 Perl (if you can't get Perl to work, send mail to the address
618 I<perlbug@perl.org> or I<perlbug@perl.com>). Reporting bugs through
619 I<perlbug> feeds into the automated bug-tracking system, access to
620 which is provided through the web at http://rt.perl.org/rt3/ . It
621 often pays to check the archives of the perl5-porters mailing list to
622 see whether the bug you're reporting has been reported before, and if
623 so whether it was considered a bug. See above for the location of
624 the searchable archives.
626 The CPAN testers ( http://testers.cpan.org/ ) are a group of
627 volunteers who test CPAN modules on a variety of platforms. Perl
628 Smokers ( http://www.nntp.perl.org/group/perl.daily-build and
629 http://www.nntp.perl.org/group/perl.daily-build.reports/ )
630 automatically test Perl source releases on platforms with various
631 configurations. Both efforts welcome volunteers. In order to get
632 involved in smoke testing of the perl itself visit
633 L<http://search.cpan.org/dist/Test-Smoke>. In order to start smoke
634 testing CPAN modules visit L<http://search.cpan.org/dist/CPAN-YACSmoke/>
635 or L<http://search.cpan.org/dist/POE-Component-CPAN-YACSmoke/> or
636 L<http://search.cpan.org/dist/CPAN-Reporter/>.
638 It's a good idea to read and lurk for a while before chipping in.
639 That way you'll get to see the dynamic of the conversations, learn the
640 personalities of the players, and hopefully be better prepared to make
641 a useful contribution when do you speak up.
643 If after all this you still think you want to join the perl5-porters
644 mailing list, send mail to I<perl5-porters-subscribe@perl.org>. To
645 unsubscribe, send mail to I<perl5-porters-unsubscribe@perl.org>.
647 To hack on the Perl guts, you'll need to read the following things:
653 This is of paramount importance, since it's the documentation of what
654 goes where in the Perl source. Read it over a couple of times and it
655 might start to make sense - don't worry if it doesn't yet, because the
656 best way to study it is to read it in conjunction with poking at Perl
657 source, and we'll do that later on.
659 You might also want to look at Gisle Aas's illustrated perlguts -
660 there's no guarantee that this will be absolutely up-to-date with the
661 latest documentation in the Perl core, but the fundamentals will be
662 right. ( http://gisle.aas.no/perl/illguts/ )
664 =item L<perlxstut> and L<perlxs>
666 A working knowledge of XSUB programming is incredibly useful for core
667 hacking; XSUBs use techniques drawn from the PP code, the portion of the
668 guts that actually executes a Perl program. It's a lot gentler to learn
669 those techniques from simple examples and explanation than from the core
674 The documentation for the Perl API explains what some of the internal
675 functions do, as well as the many macros used in the source.
677 =item F<Porting/pumpkin.pod>
679 This is a collection of words of wisdom for a Perl porter; some of it is
680 only useful to the pumpkin holder, but most of it applies to anyone
681 wanting to go about Perl development.
683 =item The perl5-porters FAQ
685 This should be available from http://dev.perl.org/perl5/docs/p5p-faq.html .
686 It contains hints on reading perl5-porters, information on how
687 perl5-porters works and how Perl development in general works.
691 =head2 Finding Your Way Around
693 Perl maintenance can be split into a number of areas, and certain people
694 (pumpkins) will have responsibility for each area. These areas sometimes
695 correspond to files or directories in the source kit. Among the areas are:
701 Modules shipped as part of the Perl core live in the F<lib/> and F<ext/>
702 subdirectories: F<lib/> is for the pure-Perl modules, and F<ext/>
703 contains the core XS modules.
707 There are tests for nearly all the modules, built-ins and major bits
708 of functionality. Test files all have a .t suffix. Module tests live
709 in the F<lib/> and F<ext/> directories next to the module being
710 tested. Others live in F<t/>. See L<Writing a test>
714 Documentation maintenance includes looking after everything in the
715 F<pod/> directory, (as well as contributing new documentation) and
716 the documentation to the modules in core.
720 The configure process is the way we make Perl portable across the
721 myriad of operating systems it supports. Responsibility for the
722 configure, build and installation process, as well as the overall
723 portability of the core code rests with the configure pumpkin - others
724 help out with individual operating systems.
726 The files involved are the operating system directories, (F<win32/>,
727 F<os2/>, F<vms/> and so on) the shell scripts which generate F<config.h>
728 and F<Makefile>, as well as the metaconfig files which generate
729 F<Configure>. (metaconfig isn't included in the core distribution.)
733 And of course, there's the core of the Perl interpreter itself. Let's
734 have a look at that in a little more detail.
738 Before we leave looking at the layout, though, don't forget that
739 F<MANIFEST> contains not only the file names in the Perl distribution,
740 but short descriptions of what's in them, too. For an overview of the
741 important files, try this:
743 perl -lne 'print if /^[^\/]+\.[ch]\s+/' MANIFEST
745 =head2 Elements of the interpreter
747 The work of the interpreter has two main stages: compiling the code
748 into the internal representation, or bytecode, and then executing it.
749 L<perlguts/Compiled code> explains exactly how the compilation stage
752 Here is a short breakdown of perl's operation:
758 The action begins in F<perlmain.c>. (or F<miniperlmain.c> for miniperl)
759 This is very high-level code, enough to fit on a single screen, and it
760 resembles the code found in L<perlembed>; most of the real action takes
763 First, F<perlmain.c> allocates some memory and constructs a Perl
766 1 PERL_SYS_INIT3(&argc,&argv,&env);
768 3 if (!PL_do_undump) {
769 4 my_perl = perl_alloc();
772 7 perl_construct(my_perl);
773 8 PL_perl_destruct_level = 0;
776 Line 1 is a macro, and its definition is dependent on your operating
777 system. Line 3 references C<PL_do_undump>, a global variable - all
778 global variables in Perl start with C<PL_>. This tells you whether the
779 current running program was created with the C<-u> flag to perl and then
780 F<undump>, which means it's going to be false in any sane context.
782 Line 4 calls a function in F<perl.c> to allocate memory for a Perl
783 interpreter. It's quite a simple function, and the guts of it looks like
786 my_perl = (PerlInterpreter*)PerlMem_malloc(sizeof(PerlInterpreter));
788 Here you see an example of Perl's system abstraction, which we'll see
789 later: C<PerlMem_malloc> is either your system's C<malloc>, or Perl's
790 own C<malloc> as defined in F<malloc.c> if you selected that option at
793 Next, in line 7, we construct the interpreter; this sets up all the
794 special variables that Perl needs, the stacks, and so on.
796 Now we pass Perl the command line options, and tell it to go:
798 exitstatus = perl_parse(my_perl, xs_init, argc, argv, (char **)NULL);
800 exitstatus = perl_run(my_perl);
804 C<perl_parse> is actually a wrapper around C<S_parse_body>, as defined
805 in F<perl.c>, which processes the command line options, sets up any
806 statically linked XS modules, opens the program and calls C<yyparse> to
811 The aim of this stage is to take the Perl source, and turn it into an op
812 tree. We'll see what one of those looks like later. Strictly speaking,
813 there's three things going on here.
815 C<yyparse>, the parser, lives in F<perly.c>, although you're better off
816 reading the original YACC input in F<perly.y>. (Yes, Virginia, there
817 B<is> a YACC grammar for Perl!) The job of the parser is to take your
818 code and "understand" it, splitting it into sentences, deciding which
819 operands go with which operators and so on.
821 The parser is nobly assisted by the lexer, which chunks up your input
822 into tokens, and decides what type of thing each token is: a variable
823 name, an operator, a bareword, a subroutine, a core function, and so on.
824 The main point of entry to the lexer is C<yylex>, and that and its
825 associated routines can be found in F<toke.c>. Perl isn't much like
826 other computer languages; it's highly context sensitive at times, it can
827 be tricky to work out what sort of token something is, or where a token
828 ends. As such, there's a lot of interplay between the tokeniser and the
829 parser, which can get pretty frightening if you're not used to it.
831 As the parser understands a Perl program, it builds up a tree of
832 operations for the interpreter to perform during execution. The routines
833 which construct and link together the various operations are to be found
834 in F<op.c>, and will be examined later.
838 Now the parsing stage is complete, and the finished tree represents
839 the operations that the Perl interpreter needs to perform to execute our
840 program. Next, Perl does a dry run over the tree looking for
841 optimisations: constant expressions such as C<3 + 4> will be computed
842 now, and the optimizer will also see if any multiple operations can be
843 replaced with a single one. For instance, to fetch the variable C<$foo>,
844 instead of grabbing the glob C<*foo> and looking at the scalar
845 component, the optimizer fiddles the op tree to use a function which
846 directly looks up the scalar in question. The main optimizer is C<peep>
847 in F<op.c>, and many ops have their own optimizing functions.
851 Now we're finally ready to go: we have compiled Perl byte code, and all
852 that's left to do is run it. The actual execution is done by the
853 C<runops_standard> function in F<run.c>; more specifically, it's done by
854 these three innocent looking lines:
856 while ((PL_op = CALL_FPTR(PL_op->op_ppaddr)(aTHX))) {
860 You may be more comfortable with the Perl version of that:
862 PERL_ASYNC_CHECK() while $Perl::op = &{$Perl::op->{function}};
864 Well, maybe not. Anyway, each op contains a function pointer, which
865 stipulates the function which will actually carry out the operation.
866 This function will return the next op in the sequence - this allows for
867 things like C<if> which choose the next op dynamically at run time.
868 The C<PERL_ASYNC_CHECK> makes sure that things like signals interrupt
869 execution if required.
871 The actual functions called are known as PP code, and they're spread
872 between four files: F<pp_hot.c> contains the "hot" code, which is most
873 often used and highly optimized, F<pp_sys.c> contains all the
874 system-specific functions, F<pp_ctl.c> contains the functions which
875 implement control structures (C<if>, C<while> and the like) and F<pp.c>
876 contains everything else. These are, if you like, the C code for Perl's
877 built-in functions and operators.
879 Note that each C<pp_> function is expected to return a pointer to the next
880 op. Calls to perl subs (and eval blocks) are handled within the same
881 runops loop, and do not consume extra space on the C stack. For example,
882 C<pp_entersub> and C<pp_entertry> just push a C<CxSUB> or C<CxEVAL> block
883 struct onto the context stack which contain the address of the op
884 following the sub call or eval. They then return the first op of that sub
885 or eval block, and so execution continues of that sub or block. Later, a
886 C<pp_leavesub> or C<pp_leavetry> op pops the C<CxSUB> or C<CxEVAL>,
887 retrieves the return op from it, and returns it.
889 =item Exception handing
891 Perl's exception handing (i.e. C<die> etc) is built on top of the low-level
892 C<setjmp()>/C<longjmp()> C-library functions. These basically provide a
893 way to capture the current PC and SP registers and later restore them; i.e.
894 a C<longjmp()> continues at the point in code where a previous C<setjmp()>
895 was done, with anything further up on the C stack being lost. This is why
896 code should always save values using C<SAVE_FOO> rather than in auto
899 The perl core wraps C<setjmp()> etc in the macros C<JMPENV_PUSH> and
900 C<JMPENV_JUMP>. The basic rule of perl exceptions is that C<exit>, and
901 C<die> (in the absence of C<eval>) perform a C<JMPENV_JUMP(2)>, while
902 C<die> within C<eval> does a C<JMPENV_JUMP(3)>.
904 At entry points to perl, such as C<perl_parse()>, C<perl_run()> and
905 C<call_sv(cv, G_EVAL)> each does a C<JMPENV_PUSH>, then enter a runops
906 loop or whatever, and handle possible exception returns. For a 2 return,
907 final cleanup is performed, such as popping stacks and calling C<CHECK> or
908 C<END> blocks. Amongst other things, this is how scope cleanup still
909 occurs during an C<exit>.
911 If a C<die> can find a C<CxEVAL> block on the context stack, then the
912 stack is popped to that level and the return op in that block is assigned
913 to C<PL_restartop>; then a C<JMPENV_JUMP(3)> is performed. This normally
914 passes control back to the guard. In the case of C<perl_run> and
915 C<call_sv>, a non-null C<PL_restartop> triggers re-entry to the runops
916 loop. The is the normal way that C<die> or C<croak> is handled within an
919 Sometimes ops are executed within an inner runops loop, such as tie, sort
920 or overload code. In this case, something like
922 sub FETCH { eval { die } }
924 would cause a longjmp right back to the guard in C<perl_run>, popping both
925 runops loops, which is clearly incorrect. One way to avoid this is for the
926 tie code to do a C<JMPENV_PUSH> before executing C<FETCH> in the inner
927 runops loop, but for efficiency reasons, perl in fact just sets a flag,
928 using C<CATCH_SET(TRUE)>. The C<pp_require>, C<pp_entereval> and
929 C<pp_entertry> ops check this flag, and if true, they call C<docatch>,
930 which does a C<JMPENV_PUSH> and starts a new runops level to execute the
931 code, rather than doing it on the current loop.
933 As a further optimisation, on exit from the eval block in the C<FETCH>,
934 execution of the code following the block is still carried on in the inner
935 loop. When an exception is raised, C<docatch> compares the C<JMPENV>
936 level of the C<CxEVAL> with C<PL_top_env> and if they differ, just
937 re-throws the exception. In this way any inner loops get popped.
941 1: eval { tie @a, 'A' };
947 To run this code, C<perl_run> is called, which does a C<JMPENV_PUSH> then
948 enters a runops loop. This loop executes the eval and tie ops on line 1,
949 with the eval pushing a C<CxEVAL> onto the context stack.
951 The C<pp_tie> does a C<CATCH_SET(TRUE)>, then starts a second runops loop
952 to execute the body of C<TIEARRAY>. When it executes the entertry op on
953 line 3, C<CATCH_GET> is true, so C<pp_entertry> calls C<docatch> which
954 does a C<JMPENV_PUSH> and starts a third runops loop, which then executes
955 the die op. At this point the C call stack looks like this:
958 Perl_runops # third loop
962 Perl_runops # second loop
966 Perl_runops # first loop
971 and the context and data stacks, as shown by C<-Dstv>, look like:
975 CX 1: EVAL => AV() PV("A"\0)
983 The die pops the first C<CxEVAL> off the context stack, sets
984 C<PL_restartop> from it, does a C<JMPENV_JUMP(3)>, and control returns to
985 the top C<docatch>. This then starts another third-level runops level,
986 which executes the nextstate, pushmark and die ops on line 4. At the point
987 that the second C<pp_die> is called, the C call stack looks exactly like
988 that above, even though we are no longer within an inner eval; this is
989 because of the optimization mentioned earlier. However, the context stack
990 now looks like this, ie with the top CxEVAL popped:
994 CX 1: EVAL => AV() PV("A"\0)
1000 The die on line 4 pops the context stack back down to the CxEVAL, leaving
1006 As usual, C<PL_restartop> is extracted from the C<CxEVAL>, and a
1007 C<JMPENV_JUMP(3)> done, which pops the C stack back to the docatch:
1011 Perl_runops # second loop
1015 Perl_runops # first loop
1020 In this case, because the C<JMPENV> level recorded in the C<CxEVAL>
1021 differs from the current one, C<docatch> just does a C<JMPENV_JUMP(3)>
1022 and the C stack unwinds to:
1027 Because C<PL_restartop> is non-null, C<run_body> starts a new runops loop
1028 and execution continues.
1032 =head2 Internal Variable Types
1034 You should by now have had a look at L<perlguts>, which tells you about
1035 Perl's internal variable types: SVs, HVs, AVs and the rest. If not, do
1038 These variables are used not only to represent Perl-space variables, but
1039 also any constants in the code, as well as some structures completely
1040 internal to Perl. The symbol table, for instance, is an ordinary Perl
1041 hash. Your code is represented by an SV as it's read into the parser;
1042 any program files you call are opened via ordinary Perl filehandles, and
1045 The core L<Devel::Peek|Devel::Peek> module lets us examine SVs from a
1046 Perl program. Let's see, for instance, how Perl treats the constant
1049 % perl -MDevel::Peek -e 'Dump("hello")'
1050 1 SV = PV(0xa041450) at 0xa04ecbc
1052 3 FLAGS = (POK,READONLY,pPOK)
1053 4 PV = 0xa0484e0 "hello"\0
1057 Reading C<Devel::Peek> output takes a bit of practise, so let's go
1058 through it line by line.
1060 Line 1 tells us we're looking at an SV which lives at C<0xa04ecbc> in
1061 memory. SVs themselves are very simple structures, but they contain a
1062 pointer to a more complex structure. In this case, it's a PV, a
1063 structure which holds a string value, at location C<0xa041450>. Line 2
1064 is the reference count; there are no other references to this data, so
1067 Line 3 are the flags for this SV - it's OK to use it as a PV, it's a
1068 read-only SV (because it's a constant) and the data is a PV internally.
1069 Next we've got the contents of the string, starting at location
1072 Line 5 gives us the current length of the string - note that this does
1073 B<not> include the null terminator. Line 6 is not the length of the
1074 string, but the length of the currently allocated buffer; as the string
1075 grows, Perl automatically extends the available storage via a routine
1078 You can get at any of these quantities from C very easily; just add
1079 C<Sv> to the name of the field shown in the snippet, and you've got a
1080 macro which will return the value: C<SvCUR(sv)> returns the current
1081 length of the string, C<SvREFCOUNT(sv)> returns the reference count,
1082 C<SvPV(sv, len)> returns the string itself with its length, and so on.
1083 More macros to manipulate these properties can be found in L<perlguts>.
1085 Let's take an example of manipulating a PV, from C<sv_catpvn>, in F<sv.c>
1088 2 Perl_sv_catpvn(pTHX_ register SV *sv, register const char *ptr, register STRLEN len)
1093 6 junk = SvPV_force(sv, tlen);
1094 7 SvGROW(sv, tlen + len + 1);
1097 10 Move(ptr,SvPVX(sv)+tlen,len,char);
1098 11 SvCUR(sv) += len;
1099 12 *SvEND(sv) = '\0';
1100 13 (void)SvPOK_only_UTF8(sv); /* validate pointer */
1104 This is a function which adds a string, C<ptr>, of length C<len> onto
1105 the end of the PV stored in C<sv>. The first thing we do in line 6 is
1106 make sure that the SV B<has> a valid PV, by calling the C<SvPV_force>
1107 macro to force a PV. As a side effect, C<tlen> gets set to the current
1108 value of the PV, and the PV itself is returned to C<junk>.
1110 In line 7, we make sure that the SV will have enough room to accommodate
1111 the old string, the new string and the null terminator. If C<LEN> isn't
1112 big enough, C<SvGROW> will reallocate space for us.
1114 Now, if C<junk> is the same as the string we're trying to add, we can
1115 grab the string directly from the SV; C<SvPVX> is the address of the PV
1118 Line 10 does the actual catenation: the C<Move> macro moves a chunk of
1119 memory around: we move the string C<ptr> to the end of the PV - that's
1120 the start of the PV plus its current length. We're moving C<len> bytes
1121 of type C<char>. After doing so, we need to tell Perl we've extended the
1122 string, by altering C<CUR> to reflect the new length. C<SvEND> is a
1123 macro which gives us the end of the string, so that needs to be a
1126 Line 13 manipulates the flags; since we've changed the PV, any IV or NV
1127 values will no longer be valid: if we have C<$a=10; $a.="6";> we don't
1128 want to use the old IV of 10. C<SvPOK_only_utf8> is a special UTF-8-aware
1129 version of C<SvPOK_only>, a macro which turns off the IOK and NOK flags
1130 and turns on POK. The final C<SvTAINT> is a macro which launders tainted
1131 data if taint mode is turned on.
1133 AVs and HVs are more complicated, but SVs are by far the most common
1134 variable type being thrown around. Having seen something of how we
1135 manipulate these, let's go on and look at how the op tree is
1140 First, what is the op tree, anyway? The op tree is the parsed
1141 representation of your program, as we saw in our section on parsing, and
1142 it's the sequence of operations that Perl goes through to execute your
1143 program, as we saw in L</Running>.
1145 An op is a fundamental operation that Perl can perform: all the built-in
1146 functions and operators are ops, and there are a series of ops which
1147 deal with concepts the interpreter needs internally - entering and
1148 leaving a block, ending a statement, fetching a variable, and so on.
1150 The op tree is connected in two ways: you can imagine that there are two
1151 "routes" through it, two orders in which you can traverse the tree.
1152 First, parse order reflects how the parser understood the code, and
1153 secondly, execution order tells perl what order to perform the
1156 The easiest way to examine the op tree is to stop Perl after it has
1157 finished parsing, and get it to dump out the tree. This is exactly what
1158 the compiler backends L<B::Terse|B::Terse>, L<B::Concise|B::Concise>
1159 and L<B::Debug|B::Debug> do.
1161 Let's have a look at how Perl sees C<$a = $b + $c>:
1163 % perl -MO=Terse -e '$a=$b+$c'
1164 1 LISTOP (0x8179888) leave
1165 2 OP (0x81798b0) enter
1166 3 COP (0x8179850) nextstate
1167 4 BINOP (0x8179828) sassign
1168 5 BINOP (0x8179800) add [1]
1169 6 UNOP (0x81796e0) null [15]
1170 7 SVOP (0x80fafe0) gvsv GV (0x80fa4cc) *b
1171 8 UNOP (0x81797e0) null [15]
1172 9 SVOP (0x8179700) gvsv GV (0x80efeb0) *c
1173 10 UNOP (0x816b4f0) null [15]
1174 11 SVOP (0x816dcf0) gvsv GV (0x80fa460) *a
1176 Let's start in the middle, at line 4. This is a BINOP, a binary
1177 operator, which is at location C<0x8179828>. The specific operator in
1178 question is C<sassign> - scalar assignment - and you can find the code
1179 which implements it in the function C<pp_sassign> in F<pp_hot.c>. As a
1180 binary operator, it has two children: the add operator, providing the
1181 result of C<$b+$c>, is uppermost on line 5, and the left hand side is on
1184 Line 10 is the null op: this does exactly nothing. What is that doing
1185 there? If you see the null op, it's a sign that something has been
1186 optimized away after parsing. As we mentioned in L</Optimization>,
1187 the optimization stage sometimes converts two operations into one, for
1188 example when fetching a scalar variable. When this happens, instead of
1189 rewriting the op tree and cleaning up the dangling pointers, it's easier
1190 just to replace the redundant operation with the null op. Originally,
1191 the tree would have looked like this:
1193 10 SVOP (0x816b4f0) rv2sv [15]
1194 11 SVOP (0x816dcf0) gv GV (0x80fa460) *a
1196 That is, fetch the C<a> entry from the main symbol table, and then look
1197 at the scalar component of it: C<gvsv> (C<pp_gvsv> into F<pp_hot.c>)
1198 happens to do both these things.
1200 The right hand side, starting at line 5 is similar to what we've just
1201 seen: we have the C<add> op (C<pp_add> also in F<pp_hot.c>) add together
1204 Now, what's this about?
1206 1 LISTOP (0x8179888) leave
1207 2 OP (0x81798b0) enter
1208 3 COP (0x8179850) nextstate
1210 C<enter> and C<leave> are scoping ops, and their job is to perform any
1211 housekeeping every time you enter and leave a block: lexical variables
1212 are tidied up, unreferenced variables are destroyed, and so on. Every
1213 program will have those first three lines: C<leave> is a list, and its
1214 children are all the statements in the block. Statements are delimited
1215 by C<nextstate>, so a block is a collection of C<nextstate> ops, with
1216 the ops to be performed for each statement being the children of
1217 C<nextstate>. C<enter> is a single op which functions as a marker.
1219 That's how Perl parsed the program, from top to bottom:
1232 However, it's impossible to B<perform> the operations in this order:
1233 you have to find the values of C<$b> and C<$c> before you add them
1234 together, for instance. So, the other thread that runs through the op
1235 tree is the execution order: each op has a field C<op_next> which points
1236 to the next op to be run, so following these pointers tells us how perl
1237 executes the code. We can traverse the tree in this order using
1238 the C<exec> option to C<B::Terse>:
1240 % perl -MO=Terse,exec -e '$a=$b+$c'
1241 1 OP (0x8179928) enter
1242 2 COP (0x81798c8) nextstate
1243 3 SVOP (0x81796c8) gvsv GV (0x80fa4d4) *b
1244 4 SVOP (0x8179798) gvsv GV (0x80efeb0) *c
1245 5 BINOP (0x8179878) add [1]
1246 6 SVOP (0x816dd38) gvsv GV (0x80fa468) *a
1247 7 BINOP (0x81798a0) sassign
1248 8 LISTOP (0x8179900) leave
1250 This probably makes more sense for a human: enter a block, start a
1251 statement. Get the values of C<$b> and C<$c>, and add them together.
1252 Find C<$a>, and assign one to the other. Then leave.
1254 The way Perl builds up these op trees in the parsing process can be
1255 unravelled by examining F<perly.y>, the YACC grammar. Let's take the
1256 piece we need to construct the tree for C<$a = $b + $c>
1258 1 term : term ASSIGNOP term
1259 2 { $$ = newASSIGNOP(OPf_STACKED, $1, $2, $3); }
1261 4 { $$ = newBINOP($2, 0, scalar($1), scalar($3)); }
1263 If you're not used to reading BNF grammars, this is how it works: You're
1264 fed certain things by the tokeniser, which generally end up in upper
1265 case. Here, C<ADDOP>, is provided when the tokeniser sees C<+> in your
1266 code. C<ASSIGNOP> is provided when C<=> is used for assigning. These are
1267 "terminal symbols", because you can't get any simpler than them.
1269 The grammar, lines one and three of the snippet above, tells you how to
1270 build up more complex forms. These complex forms, "non-terminal symbols"
1271 are generally placed in lower case. C<term> here is a non-terminal
1272 symbol, representing a single expression.
1274 The grammar gives you the following rule: you can make the thing on the
1275 left of the colon if you see all the things on the right in sequence.
1276 This is called a "reduction", and the aim of parsing is to completely
1277 reduce the input. There are several different ways you can perform a
1278 reduction, separated by vertical bars: so, C<term> followed by C<=>
1279 followed by C<term> makes a C<term>, and C<term> followed by C<+>
1280 followed by C<term> can also make a C<term>.
1282 So, if you see two terms with an C<=> or C<+>, between them, you can
1283 turn them into a single expression. When you do this, you execute the
1284 code in the block on the next line: if you see C<=>, you'll do the code
1285 in line 2. If you see C<+>, you'll do the code in line 4. It's this code
1286 which contributes to the op tree.
1289 { $$ = newBINOP($2, 0, scalar($1), scalar($3)); }
1291 What this does is creates a new binary op, and feeds it a number of
1292 variables. The variables refer to the tokens: C<$1> is the first token in
1293 the input, C<$2> the second, and so on - think regular expression
1294 backreferences. C<$$> is the op returned from this reduction. So, we
1295 call C<newBINOP> to create a new binary operator. The first parameter to
1296 C<newBINOP>, a function in F<op.c>, is the op type. It's an addition
1297 operator, so we want the type to be C<ADDOP>. We could specify this
1298 directly, but it's right there as the second token in the input, so we
1299 use C<$2>. The second parameter is the op's flags: 0 means "nothing
1300 special". Then the things to add: the left and right hand side of our
1301 expression, in scalar context.
1305 When perl executes something like C<addop>, how does it pass on its
1306 results to the next op? The answer is, through the use of stacks. Perl
1307 has a number of stacks to store things it's currently working on, and
1308 we'll look at the three most important ones here.
1312 =item Argument stack
1314 Arguments are passed to PP code and returned from PP code using the
1315 argument stack, C<ST>. The typical way to handle arguments is to pop
1316 them off the stack, deal with them how you wish, and then push the result
1317 back onto the stack. This is how, for instance, the cosine operator
1322 value = Perl_cos(value);
1325 We'll see a more tricky example of this when we consider Perl's macros
1326 below. C<POPn> gives you the NV (floating point value) of the top SV on
1327 the stack: the C<$x> in C<cos($x)>. Then we compute the cosine, and push
1328 the result back as an NV. The C<X> in C<XPUSHn> means that the stack
1329 should be extended if necessary - it can't be necessary here, because we
1330 know there's room for one more item on the stack, since we've just
1331 removed one! The C<XPUSH*> macros at least guarantee safety.
1333 Alternatively, you can fiddle with the stack directly: C<SP> gives you
1334 the first element in your portion of the stack, and C<TOP*> gives you
1335 the top SV/IV/NV/etc. on the stack. So, for instance, to do unary
1336 negation of an integer:
1340 Just set the integer value of the top stack entry to its negation.
1342 Argument stack manipulation in the core is exactly the same as it is in
1343 XSUBs - see L<perlxstut>, L<perlxs> and L<perlguts> for a longer
1344 description of the macros used in stack manipulation.
1348 I say "your portion of the stack" above because PP code doesn't
1349 necessarily get the whole stack to itself: if your function calls
1350 another function, you'll only want to expose the arguments aimed for the
1351 called function, and not (necessarily) let it get at your own data. The
1352 way we do this is to have a "virtual" bottom-of-stack, exposed to each
1353 function. The mark stack keeps bookmarks to locations in the argument
1354 stack usable by each function. For instance, when dealing with a tied
1355 variable, (internally, something with "P" magic) Perl has to call
1356 methods for accesses to the tied variables. However, we need to separate
1357 the arguments exposed to the method to the argument exposed to the
1358 original function - the store or fetch or whatever it may be. Here's
1359 roughly how the tied C<push> is implemented; see C<av_push> in F<av.c>:
1363 3 PUSHs(SvTIED_obj((SV*)av, mg));
1367 7 call_method("PUSH", G_SCALAR|G_DISCARD);
1370 Let's examine the whole implementation, for practice:
1374 Push the current state of the stack pointer onto the mark stack. This is
1375 so that when we've finished adding items to the argument stack, Perl
1376 knows how many things we've added recently.
1379 3 PUSHs(SvTIED_obj((SV*)av, mg));
1382 We're going to add two more items onto the argument stack: when you have
1383 a tied array, the C<PUSH> subroutine receives the object and the value
1384 to be pushed, and that's exactly what we have here - the tied object,
1385 retrieved with C<SvTIED_obj>, and the value, the SV C<val>.
1389 Next we tell Perl to update the global stack pointer from our internal
1390 variable: C<dSP> only gave us a local copy, not a reference to the global.
1393 7 call_method("PUSH", G_SCALAR|G_DISCARD);
1396 C<ENTER> and C<LEAVE> localise a block of code - they make sure that all
1397 variables are tidied up, everything that has been localised gets
1398 its previous value returned, and so on. Think of them as the C<{> and
1399 C<}> of a Perl block.
1401 To actually do the magic method call, we have to call a subroutine in
1402 Perl space: C<call_method> takes care of that, and it's described in
1403 L<perlcall>. We call the C<PUSH> method in scalar context, and we're
1404 going to discard its return value. The call_method() function
1405 removes the top element of the mark stack, so there is nothing for
1406 the caller to clean up.
1410 C doesn't have a concept of local scope, so perl provides one. We've
1411 seen that C<ENTER> and C<LEAVE> are used as scoping braces; the save
1412 stack implements the C equivalent of, for example:
1419 See L<perlguts/Localising Changes> for how to use the save stack.
1423 =head2 Millions of Macros
1425 One thing you'll notice about the Perl source is that it's full of
1426 macros. Some have called the pervasive use of macros the hardest thing
1427 to understand, others find it adds to clarity. Let's take an example,
1428 the code which implements the addition operator:
1432 3 dSP; dATARGET; tryAMAGICbin(add,opASSIGN);
1435 6 SETn( left + right );
1440 Every line here (apart from the braces, of course) contains a macro. The
1441 first line sets up the function declaration as Perl expects for PP code;
1442 line 3 sets up variable declarations for the argument stack and the
1443 target, the return value of the operation. Finally, it tries to see if
1444 the addition operation is overloaded; if so, the appropriate subroutine
1447 Line 5 is another variable declaration - all variable declarations start
1448 with C<d> - which pops from the top of the argument stack two NVs (hence
1449 C<nn>) and puts them into the variables C<right> and C<left>, hence the
1450 C<rl>. These are the two operands to the addition operator. Next, we
1451 call C<SETn> to set the NV of the return value to the result of adding
1452 the two values. This done, we return - the C<RETURN> macro makes sure
1453 that our return value is properly handled, and we pass the next operator
1454 to run back to the main run loop.
1456 Most of these macros are explained in L<perlapi>, and some of the more
1457 important ones are explained in L<perlxs> as well. Pay special attention
1458 to L<perlguts/Background and PERL_IMPLICIT_CONTEXT> for information on
1459 the C<[pad]THX_?> macros.
1461 =head2 The .i Targets
1463 You can expand the macros in a F<foo.c> file by saying
1467 which will expand the macros using cpp. Don't be scared by the results.
1469 =head1 SOURCE CODE STATIC ANALYSIS
1471 Various tools exist for analysing C source code B<statically>, as
1472 opposed to B<dynamically>, that is, without executing the code.
1473 It is possible to detect resource leaks, undefined behaviour, type
1474 mismatches, portability problems, code paths that would cause illegal
1475 memory accesses, and other similar problems by just parsing the C code
1476 and looking at the resulting graph, what does it tell about the
1477 execution and data flows. As a matter of fact, this is exactly
1478 how C compilers know to give warnings about dubious code.
1482 The good old C code quality inspector, C<lint>, is available in
1483 several platforms, but please be aware that there are several
1484 different implementations of it by different vendors, which means that
1485 the flags are not identical across different platforms.
1487 There is a lint variant called C<splint> (Secure Programming Lint)
1488 available from http://www.splint.org/ that should compile on any
1491 There are C<lint> and <splint> targets in Makefile, but you may have
1492 to diddle with the flags (see above).
1496 Coverity (http://www.coverity.com/) is a product similar to lint and
1497 as a testbed for their product they periodically check several open
1498 source projects, and they give out accounts to open source developers
1499 to the defect databases.
1501 =head2 cpd (cut-and-paste detector)
1503 The cpd tool detects cut-and-paste coding. If one instance of the
1504 cut-and-pasted code changes, all the other spots should probably be
1505 changed, too. Therefore such code should probably be turned into a
1506 subroutine or a macro.
1508 cpd (http://pmd.sourceforge.net/cpd.html) is part of the pmd project
1509 (http://pmd.sourceforge.net/). pmd was originally written for static
1510 analysis of Java code, but later the cpd part of it was extended to
1511 parse also C and C++.
1513 Download the pmd-bin-X.Y.zip () from the SourceForge site, extract the
1514 pmd-X.Y.jar from it, and then run that on source code thusly:
1516 java -cp pmd-X.Y.jar net.sourceforge.pmd.cpd.CPD --minimum-tokens 100 --files /some/where/src --language c > cpd.txt
1518 You may run into memory limits, in which case you should use the -Xmx option:
1524 Though much can be written about the inconsistency and coverage
1525 problems of gcc warnings (like C<-Wall> not meaning "all the
1526 warnings", or some common portability problems not being covered by
1527 C<-Wall>, or C<-ansi> and C<-pedantic> both being a poorly defined
1528 collection of warnings, and so forth), gcc is still a useful tool in
1529 keeping our coding nose clean.
1531 The C<-Wall> is by default on.
1533 The C<-ansi> (and its sidekick, C<-pedantic>) would be nice to be on
1534 always, but unfortunately they are not safe on all platforms, they can
1535 for example cause fatal conflicts with the system headers (Solaris
1536 being a prime example). If Configure C<-Dgccansipedantic> is used,
1537 the C<cflags> frontend selects C<-ansi -pedantic> for the platforms
1538 where they are known to be safe.
1540 Starting from Perl 5.9.4 the following extra flags are added:
1554 C<-Wdeclaration-after-statement>
1558 The following flags would be nice to have but they would first need
1559 their own Stygian stablemaster:
1573 C<-Wstrict-prototypes>
1577 The C<-Wtraditional> is another example of the annoying tendency of
1578 gcc to bundle a lot of warnings under one switch -- it would be
1579 impossible to deploy in practice because it would complain a lot -- but
1580 it does contain some warnings that would be beneficial to have available
1581 on their own, such as the warning about string constants inside macros
1582 containing the macro arguments: this behaved differently pre-ANSI
1583 than it does in ANSI, and some C compilers are still in transition,
1584 AIX being an example.
1586 =head2 Warnings of other C compilers
1588 Other C compilers (yes, there B<are> other C compilers than gcc) often
1589 have their "strict ANSI" or "strict ANSI with some portability extensions"
1590 modes on, like for example the Sun Workshop has its C<-Xa> mode on
1591 (though implicitly), or the DEC (these days, HP...) has its C<-std1>
1596 You can compile a special debugging version of Perl, which allows you
1597 to use the C<-D> option of Perl to tell more about what Perl is doing.
1598 But sometimes there is no alternative than to dive in with a debugger,
1599 either to see the stack trace of a core dump (very useful in a bug
1600 report), or trying to figure out what went wrong before the core dump
1601 happened, or how did we end up having wrong or unexpected results.
1603 =head2 Poking at Perl
1605 To really poke around with Perl, you'll probably want to build Perl for
1606 debugging, like this:
1608 ./Configure -d -D optimize=-g
1611 C<-g> is a flag to the C compiler to have it produce debugging
1612 information which will allow us to step through a running program,
1613 and to see in which C function we are at (without the debugging
1614 information we might see only the numerical addresses of the functions,
1615 which is not very helpful).
1617 F<Configure> will also turn on the C<DEBUGGING> compilation symbol which
1618 enables all the internal debugging code in Perl. There are a whole bunch
1619 of things you can debug with this: L<perlrun> lists them all, and the
1620 best way to find out about them is to play about with them. The most
1621 useful options are probably
1623 l Context (loop) stack processing
1625 o Method and overloading resolution
1626 c String/numeric conversions
1628 Some of the functionality of the debugging code can be achieved using XS
1631 -Dr => use re 'debug'
1632 -Dx => use O 'Debug'
1634 =head2 Using a source-level debugger
1636 If the debugging output of C<-D> doesn't help you, it's time to step
1637 through perl's execution with a source-level debugger.
1643 We'll use C<gdb> for our examples here; the principles will apply to
1644 any debugger (many vendors call their debugger C<dbx>), but check the
1645 manual of the one you're using.
1649 To fire up the debugger, type
1653 Or if you have a core dump:
1657 You'll want to do that in your Perl source tree so the debugger can read
1658 the source code. You should see the copyright message, followed by the
1663 C<help> will get you into the documentation, but here are the most
1670 Run the program with the given arguments.
1672 =item break function_name
1674 =item break source.c:xxx
1676 Tells the debugger that we'll want to pause execution when we reach
1677 either the named function (but see L<perlguts/Internal Functions>!) or the given
1678 line in the named source file.
1682 Steps through the program a line at a time.
1686 Steps through the program a line at a time, without descending into
1691 Run until the next breakpoint.
1695 Run until the end of the current function, then stop again.
1699 Just pressing Enter will do the most recent operation again - it's a
1700 blessing when stepping through miles of source code.
1704 Execute the given C code and print its results. B<WARNING>: Perl makes
1705 heavy use of macros, and F<gdb> does not necessarily support macros
1706 (see later L</"gdb macro support">). You'll have to substitute them
1707 yourself, or to invoke cpp on the source code files
1708 (see L</"The .i Targets">)
1709 So, for instance, you can't say
1711 print SvPV_nolen(sv)
1715 print Perl_sv_2pv_nolen(sv)
1719 You may find it helpful to have a "macro dictionary", which you can
1720 produce by saying C<cpp -dM perl.c | sort>. Even then, F<cpp> won't
1721 recursively apply those macros for you.
1723 =head2 gdb macro support
1725 Recent versions of F<gdb> have fairly good macro support, but
1726 in order to use it you'll need to compile perl with macro definitions
1727 included in the debugging information. Using F<gcc> version 3.1, this
1728 means configuring with C<-Doptimize=-g3>. Other compilers might use a
1729 different switch (if they support debugging macros at all).
1731 =head2 Dumping Perl Data Structures
1733 One way to get around this macro hell is to use the dumping functions in
1734 F<dump.c>; these work a little like an internal
1735 L<Devel::Peek|Devel::Peek>, but they also cover OPs and other structures
1736 that you can't get at from Perl. Let's take an example. We'll use the
1737 C<$a = $b + $c> we used before, but give it a bit of context:
1738 C<$b = "6XXXX"; $c = 2.3;>. Where's a good place to stop and poke around?
1740 What about C<pp_add>, the function we examined earlier to implement the
1743 (gdb) break Perl_pp_add
1744 Breakpoint 1 at 0x46249f: file pp_hot.c, line 309.
1746 Notice we use C<Perl_pp_add> and not C<pp_add> - see L<perlguts/Internal Functions>.
1747 With the breakpoint in place, we can run our program:
1749 (gdb) run -e '$b = "6XXXX"; $c = 2.3; $a = $b + $c'
1751 Lots of junk will go past as gdb reads in the relevant source files and
1752 libraries, and then:
1754 Breakpoint 1, Perl_pp_add () at pp_hot.c:309
1755 309 dSP; dATARGET; tryAMAGICbin(add,opASSIGN);
1760 We looked at this bit of code before, and we said that C<dPOPTOPnnrl_ul>
1761 arranges for two C<NV>s to be placed into C<left> and C<right> - let's
1764 #define dPOPTOPnnrl_ul NV right = POPn; \
1765 SV *leftsv = TOPs; \
1766 NV left = USE_LEFT(leftsv) ? SvNV(leftsv) : 0.0
1768 C<POPn> takes the SV from the top of the stack and obtains its NV either
1769 directly (if C<SvNOK> is set) or by calling the C<sv_2nv> function.
1770 C<TOPs> takes the next SV from the top of the stack - yes, C<POPn> uses
1771 C<TOPs> - but doesn't remove it. We then use C<SvNV> to get the NV from
1772 C<leftsv> in the same way as before - yes, C<POPn> uses C<SvNV>.
1774 Since we don't have an NV for C<$b>, we'll have to use C<sv_2nv> to
1775 convert it. If we step again, we'll find ourselves there:
1777 Perl_sv_2nv (sv=0xa0675d0) at sv.c:1669
1781 We can now use C<Perl_sv_dump> to investigate the SV:
1783 SV = PV(0xa057cc0) at 0xa0675d0
1786 PV = 0xa06a510 "6XXXX"\0
1791 We know we're going to get C<6> from this, so let's finish the
1795 Run till exit from #0 Perl_sv_2nv (sv=0xa0675d0) at sv.c:1671
1796 0x462669 in Perl_pp_add () at pp_hot.c:311
1799 We can also dump out this op: the current op is always stored in
1800 C<PL_op>, and we can dump it with C<Perl_op_dump>. This'll give us
1801 similar output to L<B::Debug|B::Debug>.
1804 13 TYPE = add ===> 14
1806 FLAGS = (SCALAR,KIDS)
1808 TYPE = null ===> (12)
1810 FLAGS = (SCALAR,KIDS)
1812 11 TYPE = gvsv ===> 12
1818 # finish this later #
1822 All right, we've now had a look at how to navigate the Perl sources and
1823 some things you'll need to know when fiddling with them. Let's now get
1824 on and create a simple patch. Here's something Larry suggested: if a
1825 C<U> is the first active format during a C<pack>, (for example,
1826 C<pack "U3C8", @stuff>) then the resulting string should be treated as
1829 How do we prepare to fix this up? First we locate the code in question -
1830 the C<pack> happens at runtime, so it's going to be in one of the F<pp>
1831 files. Sure enough, C<pp_pack> is in F<pp.c>. Since we're going to be
1832 altering this file, let's copy it to F<pp.c~>.
1834 [Well, it was in F<pp.c> when this tutorial was written. It has now been
1835 split off with C<pp_unpack> to its own file, F<pp_pack.c>]
1837 Now let's look over C<pp_pack>: we take a pattern into C<pat>, and then
1838 loop over the pattern, taking each format character in turn into
1839 C<datum_type>. Then for each possible format character, we swallow up
1840 the other arguments in the pattern (a field width, an asterisk, and so
1841 on) and convert the next chunk input into the specified format, adding
1842 it onto the output SV C<cat>.
1844 How do we know if the C<U> is the first format in the C<pat>? Well, if
1845 we have a pointer to the start of C<pat> then, if we see a C<U> we can
1846 test whether we're still at the start of the string. So, here's where
1850 register char *pat = SvPVx(*++MARK, fromlen);
1851 register char *patend = pat + fromlen;
1856 We'll have another string pointer in there:
1859 register char *pat = SvPVx(*++MARK, fromlen);
1860 register char *patend = pat + fromlen;
1866 And just before we start the loop, we'll set C<patcopy> to be the start
1871 sv_setpvn(cat, "", 0);
1873 while (pat < patend) {
1875 Now if we see a C<U> which was at the start of the string, we turn on
1876 the C<UTF8> flag for the output SV, C<cat>:
1878 + if (datumtype == 'U' && pat==patcopy+1)
1880 if (datumtype == '#') {
1881 while (pat < patend && *pat != '\n')
1884 Remember that it has to be C<patcopy+1> because the first character of
1885 the string is the C<U> which has been swallowed into C<datumtype!>
1887 Oops, we forgot one thing: what if there are spaces at the start of the
1888 pattern? C<pack(" U*", @stuff)> will have C<U> as the first active
1889 character, even though it's not the first thing in the pattern. In this
1890 case, we have to advance C<patcopy> along with C<pat> when we see spaces:
1892 if (isSPACE(datumtype))
1897 if (isSPACE(datumtype)) {
1902 OK. That's the C part done. Now we must do two additional things before
1903 this patch is ready to go: we've changed the behaviour of Perl, and so
1904 we must document that change. We must also provide some more regression
1905 tests to make sure our patch works and doesn't create a bug somewhere
1906 else along the line.
1908 The regression tests for each operator live in F<t/op/>, and so we
1909 make a copy of F<t/op/pack.t> to F<t/op/pack.t~>. Now we can add our
1910 tests to the end. First, we'll test that the C<U> does indeed create
1913 t/op/pack.t has a sensible ok() function, but if it didn't we could
1914 use the one from t/test.pl.
1916 require './test.pl';
1917 plan( tests => 159 );
1921 print 'not ' unless "1.20.300.4000" eq sprintf "%vd", pack("U*",1,20,300,4000);
1922 print "ok $test\n"; $test++;
1924 we can write the more sensible (see L<Test::More> for a full
1925 explanation of is() and other testing functions).
1927 is( "1.20.300.4000", sprintf "%vd", pack("U*",1,20,300,4000),
1928 "U* produces Unicode" );
1930 Now we'll test that we got that space-at-the-beginning business right:
1932 is( "1.20.300.4000", sprintf "%vd", pack(" U*",1,20,300,4000),
1933 " with spaces at the beginning" );
1935 And finally we'll test that we don't make Unicode strings if C<U> is B<not>
1936 the first active format:
1938 isnt( v1.20.300.4000, sprintf "%vd", pack("C0U*",1,20,300,4000),
1939 "U* not first isn't Unicode" );
1941 Mustn't forget to change the number of tests which appears at the top,
1942 or else the automated tester will get confused. This will either look
1949 plan( tests => 156 );
1951 We now compile up Perl, and run it through the test suite. Our new
1954 Finally, the documentation. The job is never done until the paperwork is
1955 over, so let's describe the change we've just made. The relevant place
1956 is F<pod/perlfunc.pod>; again, we make a copy, and then we'll insert
1957 this text in the description of C<pack>:
1961 If the pattern begins with a C<U>, the resulting string will be treated
1962 as UTF-8-encoded Unicode. You can force UTF-8 encoding on in a string
1963 with an initial C<U0>, and the bytes that follow will be interpreted as
1964 Unicode characters. If you don't want this to happen, you can begin your
1965 pattern with C<C0> (or anything else) to force Perl not to UTF-8 encode your
1966 string, and then follow this with a C<U*> somewhere in your pattern.
1968 All done. Now let's create the patch. F<Porting/patching.pod> tells us
1969 that if we're making major changes, we should copy the entire directory
1970 to somewhere safe before we begin fiddling, and then do
1972 diff -ruN old new > patch
1974 However, we know which files we've changed, and we can simply do this:
1976 diff -u pp.c~ pp.c > patch
1977 diff -u t/op/pack.t~ t/op/pack.t >> patch
1978 diff -u pod/perlfunc.pod~ pod/perlfunc.pod >> patch
1980 We end up with a patch looking a little like this:
1982 --- pp.c~ Fri Jun 02 04:34:10 2000
1983 +++ pp.c Fri Jun 16 11:37:25 2000
1984 @@ -4375,6 +4375,7 @@
1987 register char *pat = SvPVx(*++MARK, fromlen);
1989 register char *patend = pat + fromlen;
1992 @@ -4405,6 +4406,7 @@
1995 And finally, we submit it, with our rationale, to perl5-porters. Job
1998 =head2 Patching a core module
2000 This works just like patching anything else, with an extra
2001 consideration. Many core modules also live on CPAN. If this is so,
2002 patch the CPAN version instead of the core and send the patch off to
2003 the module maintainer (with a copy to p5p). This will help the module
2004 maintainer keep the CPAN version in sync with the core version without
2005 constantly scanning p5p.
2007 The list of maintainers of core modules is usefully documented in
2008 F<Porting/Maintainers.pl>.
2010 =head2 Adding a new function to the core
2012 If, as part of a patch to fix a bug, or just because you have an
2013 especially good idea, you decide to add a new function to the core,
2014 discuss your ideas on p5p well before you start work. It may be that
2015 someone else has already attempted to do what you are considering and
2016 can give lots of good advice or even provide you with bits of code
2017 that they already started (but never finished).
2019 You have to follow all of the advice given above for patching. It is
2020 extremely important to test any addition thoroughly and add new tests
2021 to explore all boundary conditions that your new function is expected
2022 to handle. If your new function is used only by one module (e.g. toke),
2023 then it should probably be named S_your_function (for static); on the
2024 other hand, if you expect it to accessible from other functions in
2025 Perl, you should name it Perl_your_function. See L<perlguts/Internal Functions>
2028 The location of any new code is also an important consideration. Don't
2029 just create a new top level .c file and put your code there; you would
2030 have to make changes to Configure (so the Makefile is created properly),
2031 as well as possibly lots of include files. This is strictly pumpking
2034 It is better to add your function to one of the existing top level
2035 source code files, but your choice is complicated by the nature of
2036 the Perl distribution. Only the files that are marked as compiled
2037 static are located in the perl executable. Everything else is located
2038 in the shared library (or DLL if you are running under WIN32). So,
2039 for example, if a function was only used by functions located in
2040 toke.c, then your code can go in toke.c. If, however, you want to call
2041 the function from universal.c, then you should put your code in another
2042 location, for example util.c.
2044 In addition to writing your c-code, you will need to create an
2045 appropriate entry in embed.pl describing your function, then run
2046 'make regen_headers' to create the entries in the numerous header
2047 files that perl needs to compile correctly. See L<perlguts/Internal Functions>
2048 for information on the various options that you can set in embed.pl.
2049 You will forget to do this a few (or many) times and you will get
2050 warnings during the compilation phase. Make sure that you mention
2051 this when you post your patch to P5P; the pumpking needs to know this.
2053 When you write your new code, please be conscious of existing code
2054 conventions used in the perl source files. See L<perlstyle> for
2055 details. Although most of the guidelines discussed seem to focus on
2056 Perl code, rather than c, they all apply (except when they don't ;).
2057 See also I<Porting/patching.pod> file in the Perl source distribution
2058 for lots of details about both formatting and submitting patches of
2061 Lastly, TEST TEST TEST TEST TEST any code before posting to p5p.
2062 Test on as many platforms as you can find. Test as many perl
2063 Configure options as you can (e.g. MULTIPLICITY). If you have
2064 profiling or memory tools, see L<EXTERNAL TOOLS FOR DEBUGGING PERL>
2065 below for how to use them to further test your code. Remember that
2066 most of the people on P5P are doing this on their own time and
2067 don't have the time to debug your code.
2069 =head2 Writing a test
2071 Every module and built-in function has an associated test file (or
2072 should...). If you add or change functionality, you have to write a
2073 test. If you fix a bug, you have to write a test so that bug never
2074 comes back. If you alter the docs, it would be nice to test what the
2075 new documentation says.
2077 In short, if you submit a patch you probably also have to patch the
2080 For modules, the test file is right next to the module itself.
2081 F<lib/strict.t> tests F<lib/strict.pm>. This is a recent innovation,
2082 so there are some snags (and it would be wonderful for you to brush
2083 them out), but it basically works that way. Everything else lives in
2090 Testing of the absolute basic functionality of Perl. Things like
2091 C<if>, basic file reads and writes, simple regexes, etc. These are
2092 run first in the test suite and if any of them fail, something is
2097 These test the basic control structures, C<if/else>, C<while>,
2102 Tests basic issues of how Perl parses and compiles itself.
2106 Tests for built-in IO functions, including command line arguments.
2110 The old home for the module tests, you shouldn't put anything new in
2111 here. There are still some bits and pieces hanging around in here
2112 that need to be moved. Perhaps you could move them? Thanks!
2116 Tests for perl's built in functions that don't fit into any of the
2121 Tests for POD directives. There are still some tests for the Pod
2122 modules hanging around in here that need to be moved out into F<lib/>.
2126 Testing features of how perl actually runs, including exit codes and
2127 handling of PERL* environment variables.
2131 Tests for the core support of Unicode.
2135 Windows-specific tests.
2139 A test suite for the s2p converter.
2143 The core uses the same testing style as the rest of Perl, a simple
2144 "ok/not ok" run through Test::Harness, but there are a few special
2147 There are three ways to write a test in the core. Test::More,
2148 t/test.pl and ad hoc C<print $test ? "ok 42\n" : "not ok 42\n">. The
2149 decision of which to use depends on what part of the test suite you're
2150 working on. This is a measure to prevent a high-level failure (such
2151 as Config.pm breaking) from causing basic functionality tests to fail.
2157 Since we don't know if require works, or even subroutines, use ad hoc
2158 tests for these two. Step carefully to avoid using the feature being
2161 =item t/cmd t/run t/io t/op
2163 Now that basic require() and subroutines are tested, you can use the
2164 t/test.pl library which emulates the important features of Test::More
2165 while using a minimum of core features.
2167 You can also conditionally use certain libraries like Config, but be
2168 sure to skip the test gracefully if it's not there.
2172 Now that the core of Perl is tested, Test::More can be used. You can
2173 also use the full suite of core modules in the tests.
2177 When you say "make test" Perl uses the F<t/TEST> program to run the
2178 test suite (except under Win32 where it uses F<t/harness> instead.)
2179 All tests are run from the F<t/> directory, B<not> the directory
2180 which contains the test. This causes some problems with the tests
2181 in F<lib/>, so here's some opportunity for some patching.
2183 You must be triply conscious of cross-platform concerns. This usually
2184 boils down to using File::Spec and avoiding things like C<fork()> and
2185 C<system()> unless absolutely necessary.
2187 =head2 Special Make Test Targets
2189 There are various special make targets that can be used to test Perl
2190 slightly differently than the standard "test" target. Not all them
2191 are expected to give a 100% success rate. Many of them have several
2192 aliases, and many of them are not available on certain operating
2199 Run F<perl> on all core tests (F<t/*> and F<lib/[a-z]*> pragma tests).
2201 (Not available on Win32)
2205 Run all the tests through B::Deparse. Not all tests will succeed.
2207 (Not available on Win32)
2209 =item test.taintwarn
2211 Run all tests with the B<-t> command-line switch. Not all tests
2212 are expected to succeed (until they're specifically fixed, of course).
2214 (Not available on Win32)
2218 Run F<miniperl> on F<t/base>, F<t/comp>, F<t/cmd>, F<t/run>, F<t/io>,
2219 F<t/op>, and F<t/uni> tests.
2221 =item test.valgrind check.valgrind utest.valgrind ucheck.valgrind
2223 (Only in Linux) Run all the tests using the memory leak + naughty
2224 memory access tool "valgrind". The log files will be named
2225 F<testname.valgrind>.
2227 =item test.third check.third utest.third ucheck.third
2229 (Only in Tru64) Run all the tests using the memory leak + naughty
2230 memory access tool "Third Degree". The log files will be named
2231 F<perl.3log.testname>.
2233 =item test.torture torturetest
2235 Run all the usual tests and some extra tests. As of Perl 5.8.0 the
2236 only extra tests are Abigail's JAPHs, F<t/japh/abigail.t>.
2238 You can also run the torture test with F<t/harness> by giving
2239 C<-torture> argument to F<t/harness>.
2241 =item utest ucheck test.utf8 check.utf8
2243 Run all the tests with -Mutf8. Not all tests will succeed.
2245 (Not available on Win32)
2247 =item minitest.utf16 test.utf16
2249 Runs the tests with UTF-16 encoded scripts, encoded with different
2250 versions of this encoding.
2252 C<make utest.utf16> runs the test suite with a combination of C<-utf8> and
2253 C<-utf16> arguments to F<t/TEST>.
2255 (Not available on Win32)
2259 Run the test suite with the F<t/harness> controlling program, instead of
2260 F<t/TEST>. F<t/harness> is more sophisticated, and uses the
2261 L<Test::Harness> module, thus using this test target supposes that perl
2262 mostly works. The main advantage for our purposes is that it prints a
2263 detailed summary of failed tests at the end. Also, unlike F<t/TEST>, it
2264 doesn't redirect stderr to stdout.
2266 Note that under Win32 F<t/harness> is always used instead of F<t/TEST>, so
2267 there is no special "test_harness" target.
2269 Under Win32's "test" target you may use the TEST_SWITCHES and TEST_FILES
2270 environment variables to control the behaviour of F<t/harness>. This means
2273 nmake test TEST_FILES="op/*.t"
2274 nmake test TEST_SWITCHES="-torture" TEST_FILES="op/*.t"
2276 =item test-notty test_notty
2278 Sets PERL_SKIP_TTY_TEST to true before running normal test.
2282 =head2 Running tests by hand
2284 You can run part of the test suite by hand by using one the following
2285 commands from the F<t/> directory :
2287 ./perl -I../lib TEST list-of-.t-files
2291 ./perl -I../lib harness list-of-.t-files
2293 (if you don't specify test scripts, the whole test suite will be run.)
2295 =head3 Using t/harness for testing
2297 If you use C<harness> for testing you have several command line options
2298 available to you. The arguments are as follows, and are in the order
2299 that they must appear if used together.
2301 harness -v -torture -re=pattern LIST OF FILES TO TEST
2302 harness -v -torture -re LIST OF PATTERNS TO MATCH
2304 If C<LIST OF FILES TO TEST> is omitted the file list is obtained from
2305 the manifest. The file list may include shell wildcards which will be
2312 Run the tests under verbose mode so you can see what tests were run,
2317 Run the torture tests as well as the normal set.
2321 Filter the file list so that all the test files run match PATTERN.
2322 Note that this form is distinct from the B<-re LIST OF PATTERNS> form below
2323 in that it allows the file list to be provided as well.
2325 =item -re LIST OF PATTERNS
2327 Filter the file list so that all the test files run match
2328 /(LIST|OF|PATTERNS)/. Note that with this form the patterns
2329 are joined by '|' and you cannot supply a list of files, instead
2330 the test files are obtained from the MANIFEST.
2334 You can run an individual test by a command similar to
2336 ./perl -I../lib patho/to/foo.t
2338 except that the harnesses set up some environment variables that may
2339 affect the execution of the test :
2345 indicates that we're running this test part of the perl core test suite.
2346 This is useful for modules that have a dual life on CPAN.
2348 =item PERL_DESTRUCT_LEVEL=2
2350 is set to 2 if it isn't set already (see L</PERL_DESTRUCT_LEVEL>)
2354 (used only by F<t/TEST>) if set, overrides the path to the perl executable
2355 that should be used to run the tests (the default being F<./perl>).
2357 =item PERL_SKIP_TTY_TEST
2359 if set, tells to skip the tests that need a terminal. It's actually set
2360 automatically by the Makefile, but can also be forced artificially by
2361 running 'make test_notty'.
2365 =head3 Other environment variables that may influence tests
2369 =item PERL_TEST_Net_Ping
2371 Setting this variable runs all the Net::Ping modules tests,
2372 otherwise some tests that interact with the outside world are skipped.
2375 =item PERL_TEST_NOVREXX
2377 Setting this variable skips the vrexx.t tests for OS2::REXX.
2379 =item PERL_TEST_NUMCONVERTS
2381 This sets a variable in op/numconvert.t.
2385 See also the documentation for the Test and Test::Harness modules,
2386 for more environment variables that affect testing.
2388 =head2 Common problems when patching Perl source code
2390 Perl source plays by ANSI C89 rules: no C99 (or C++) extensions. In
2391 some cases we have to take pre-ANSI requirements into consideration.
2392 You don't care about some particular platform having broken Perl?
2393 I hear there is still a strong demand for J2EE programmers.
2395 =head2 Perl environment problems
2401 Not compiling with threading
2403 Compiling with threading (-Duseithreads) completely rewrites
2404 the function prototypes of Perl. You better try your changes
2405 with that. Related to this is the difference between "Perl_-less"
2406 and "Perl_-ly" APIs, for example:
2408 Perl_sv_setiv(aTHX_ ...);
2411 The first one explicitly passes in the context, which is needed for e.g.
2412 threaded builds. The second one does that implicitly; do not get them
2413 mixed. If you are not passing in a aTHX_, you will need to do a dTHX
2414 (or a dVAR) as the first thing in the function.
2416 See L<perlguts/"How multiple interpreters and concurrency are supported">
2417 for further discussion about context.
2421 Not compiling with -DDEBUGGING
2423 The DEBUGGING define exposes more code to the compiler,
2424 therefore more ways for things to go wrong. You should try it.
2428 Introducing (non-read-only) globals
2430 Do not introduce any modifiable globals, truly global or file static.
2431 They are bad form and complicate multithreading and other forms of
2432 concurrency. The right way is to introduce them as new interpreter
2433 variables, see F<intrpvar.h> (at the very end for binary compatibility).
2435 Introducing read-only (const) globals is okay, as long as you verify
2436 with e.g. C<nm libperl.a|egrep -v ' [TURtr] '> (if your C<nm> has
2437 BSD-style output) that the data you added really is read-only.
2438 (If it is, it shouldn't show up in the output of that command.)
2440 If you want to have static strings, make them constant:
2442 static const char etc[] = "...";
2444 If you want to have arrays of constant strings, note carefully
2445 the right combination of C<const>s:
2447 static const char * const yippee[] =
2448 {"hi", "ho", "silver"};
2450 There is a way to completely hide any modifiable globals (they are all
2451 moved to heap), the compilation setting C<-DPERL_GLOBAL_STRUCT_PRIVATE>.
2452 It is not normally used, but can be used for testing, read more
2453 about it in L<perlguts/"Background and PERL_IMPLICIT_CONTEXT">.
2457 Not exporting your new function
2459 Some platforms (Win32, AIX, VMS, OS/2, to name a few) require any
2460 function that is part of the public API (the shared Perl library)
2461 to be explicitly marked as exported. See the discussion about
2462 F<embed.pl> in L<perlguts>.
2466 Exporting your new function
2468 The new shiny result of either genuine new functionality or your
2469 arduous refactoring is now ready and correctly exported. So what
2470 could possibly go wrong?
2472 Maybe simply that your function did not need to be exported in the
2473 first place. Perl has a long and not so glorious history of exporting
2474 functions that it should not have.
2476 If the function is used only inside one source code file, make it
2477 static. See the discussion about F<embed.pl> in L<perlguts>.
2479 If the function is used across several files, but intended only for
2480 Perl's internal use (and this should be the common case), do not
2481 export it to the public API. See the discussion about F<embed.pl>
2486 =head2 Portability problems
2488 The following are common causes of compilation and/or execution
2489 failures, not common to Perl as such. The C FAQ is good bedtime
2490 reading. Please test your changes with as many C compilers and
2491 platforms as possible -- we will, anyway, and it's nice to save
2492 oneself from public embarrassment.
2494 If using gcc, you can add the C<-std=c89> option which will hopefully
2495 catch most of these unportabilities. (However it might also catch
2496 incompatibilities in your system's header files.)
2498 Use the Configure C<-Dgccansipedantic> flag to enable the gcc
2499 C<-ansi -pedantic> flags which enforce stricter ANSI rules.
2501 If using the C<gcc -Wall> note that not all the possible warnings
2502 (like C<-Wunitialized>) are given unless you also compile with C<-O>.
2504 Note that if using gcc, starting from Perl 5.9.5 the Perl core source
2505 code files (the ones at the top level of the source code distribution,
2506 but not e.g. the extensions under ext/) are automatically compiled
2507 with as many as possible of the C<-std=c89>, C<-ansi>, C<-pedantic>,
2508 and a selection of C<-W> flags (see cflags.SH).
2510 Also study L<perlport> carefully to avoid any bad assumptions
2511 about the operating system, filesystems, and so forth.
2513 You may once in a while try a "make microperl" to see whether we
2514 can still compile Perl with just the bare minimum of interfaces.
2517 Do not assume an operating system indicates a certain compiler.
2523 Casting pointers to integers or casting integers to pointers
2525 void castaway(U8* p)
2531 void castaway(U8* p)
2535 Both are bad, and broken, and unportable. Use the PTR2IV()
2536 macro that does it right. (Likewise, there are PTR2UV(), PTR2NV(),
2537 INT2PTR(), and NUM2PTR().)
2541 Casting between data function pointers and data pointers
2543 Technically speaking casting between function pointers and data
2544 pointers is unportable and undefined, but practically speaking
2545 it seems to work, but you should use the FPTR2DPTR() and DPTR2FPTR()
2546 macros. Sometimes you can also play games with unions.
2550 Assuming sizeof(int) == sizeof(long)
2552 There are platforms where longs are 64 bits, and platforms where ints
2553 are 64 bits, and while we are out to shock you, even platforms where
2554 shorts are 64 bits. This is all legal according to the C standard.
2555 (In other words, "long long" is not a portable way to specify 64 bits,
2556 and "long long" is not even guaranteed to be any wider than "long".)
2558 Instead, use the definitions IV, UV, IVSIZE, I32SIZE, and so forth.
2559 Avoid things like I32 because they are B<not> guaranteed to be
2560 I<exactly> 32 bits, they are I<at least> 32 bits, nor are they
2561 guaranteed to be B<int> or B<long>. If you really explicitly need
2562 64-bit variables, use I64 and U64, but only if guarded by HAS_QUAD.
2566 Assuming one can dereference any type of pointer for any type of data
2569 long pony = *p; /* BAD */
2571 Many platforms, quite rightly so, will give you a core dump instead
2572 of a pony if the p happens not be correctly aligned.
2578 (int)*p = ...; /* BAD */
2580 Simply not portable. Get your lvalue to be of the right type,
2581 or maybe use temporary variables, or dirty tricks with unions.
2585 Assume B<anything> about structs (especially the ones you
2586 don't control, like the ones coming from the system headers)
2592 That a certain field exists in a struct
2596 That no other fields exist besides the ones you know of
2600 That a field is of certain signedness, sizeof, or type
2604 That the fields are in a certain order
2610 While C guarantees the ordering specified in the struct definition,
2611 between different platforms the definitions might differ
2617 That the sizeof(struct) or the alignments are the same everywhere
2623 There might be padding bytes between the fields to align the fields -
2624 the bytes can be anything
2628 Structs are required to be aligned to the maximum alignment required
2629 by the fields - which for native types is for usually equivalent to
2630 sizeof() of the field
2638 Mixing #define and #ifdef
2640 #define BURGLE(x) ... \
2641 #ifdef BURGLE_OLD_STYLE /* BAD */
2642 ... do it the old way ... \
2644 ... do it the new way ... \
2647 You cannot portably "stack" cpp directives. For example in the above
2648 you need two separate BURGLE() #defines, one for each #ifdef branch.
2652 Adding stuff after #endif or #else
2656 #else !SNOSH /* BAD */
2658 #endif SNOSH /* BAD */
2660 The #endif and #else cannot portably have anything non-comment after
2661 them. If you want to document what is going (which is a good idea
2662 especially if the branches are long), use (C) comments:
2670 The gcc option C<-Wendif-labels> warns about the bad variant
2671 (by default on starting from Perl 5.9.4).
2675 Having a comma after the last element of an enum list
2683 is not portable. Leave out the last comma.
2685 Also note that whether enums are implicitly morphable to ints
2686 varies between compilers, you might need to (int).
2692 // This function bamfoodles the zorklator. /* BAD */
2694 That is C99 or C++. Perl is C89. Using the //-comments is silently
2695 allowed by many C compilers but cranking up the ANSI C89 strictness
2696 (which we like to do) causes the compilation to fail.
2700 Mixing declarations and code
2705 set_zorkmids(n); /* BAD */
2708 That is C99 or C++. Some C compilers allow that, but you shouldn't.
2710 The gcc option C<-Wdeclaration-after-statements> scans for such problems
2711 (by default on starting from Perl 5.9.4).
2715 Introducing variables inside for()
2717 for(int i = ...; ...; ...) { /* BAD */
2719 That is C99 or C++. While it would indeed be awfully nice to have that
2720 also in C89, to limit the scope of the loop variable, alas, we cannot.
2724 Mixing signed char pointers with unsigned char pointers
2726 int foo(char *s) { ... }
2728 unsigned char *t = ...; /* Or U8* t = ... */
2731 While this is legal practice, it is certainly dubious, and downright
2732 fatal in at least one platform: for example VMS cc considers this a
2733 fatal error. One cause for people often making this mistake is that a
2734 "naked char" and therefore dereferencing a "naked char pointer" have
2735 an undefined signedness: it depends on the compiler and the flags of
2736 the compiler and the underlying platform whether the result is signed
2737 or unsigned. For this very same reason using a 'char' as an array
2742 Macros that have string constants and their arguments as substrings of
2743 the string constants
2745 #define FOO(n) printf("number = %d\n", n) /* BAD */
2748 Pre-ANSI semantics for that was equivalent to
2750 printf("10umber = %d\10");
2752 which is probably not what you were expecting. Unfortunately at least
2753 one reasonably common and modern C compiler does "real backward
2754 compatibility" here, in AIX that is what still happens even though the
2755 rest of the AIX compiler is very happily C89.
2759 Using printf formats for non-basic C types
2762 printf("i = %d\n", i); /* BAD */
2764 While this might by accident work in some platform (where IV happens
2765 to be an C<int>), in general it cannot. IV might be something larger.
2766 Even worse the situation is with more specific types (defined by Perl's
2767 configuration step in F<config.h>):
2770 printf("who = %d\n", who); /* BAD */
2772 The problem here is that Uid_t might be not only not C<int>-wide
2773 but it might also be unsigned, in which case large uids would be
2774 printed as negative values.
2776 There is no simple solution to this because of printf()'s limited
2777 intelligence, but for many types the right format is available as
2778 with either 'f' or '_f' suffix, for example:
2780 IVdf /* IV in decimal */
2781 UVxf /* UV is hexadecimal */
2783 printf("i = %"IVdf"\n", i); /* The IVdf is a string constant. */
2785 Uid_t_f /* Uid_t in decimal */
2787 printf("who = %"Uid_t_f"\n", who);
2789 Or you can try casting to a "wide enough" type:
2791 printf("i = %"IVdf"\n", (IV)something_very_small_and_signed);
2793 Also remember that the C<%p> format really does require a void pointer:
2796 printf("p = %p\n", (void*)p);
2798 The gcc option C<-Wformat> scans for such problems.
2802 Blindly using variadic macros
2804 gcc has had them for a while with its own syntax, and C99 brought
2805 them with a standardized syntax. Don't use the former, and use
2806 the latter only if the HAS_C99_VARIADIC_MACROS is defined.
2810 Blindly passing va_list
2812 Not all platforms support passing va_list to further varargs (stdarg)
2813 functions. The right thing to do is to copy the va_list using the
2814 Perl_va_copy() if the NEED_VA_COPY is defined.
2818 Using gcc statement expressions
2820 val = ({...;...;...}); /* BAD */
2822 While a nice extension, it's not portable. The Perl code does
2823 admittedly use them if available to gain some extra speed
2824 (essentially as a funky form of inlining), but you shouldn't.
2828 Binding together several statements
2830 Use the macros STMT_START and STMT_END.
2838 Testing for operating systems or versions when should be testing for features
2840 #ifdef __FOONIX__ /* BAD */
2844 Unless you know with 100% certainty that quux() is only ever available
2845 for the "Foonix" operating system B<and> that is available B<and>
2846 correctly working for B<all> past, present, B<and> future versions of
2847 "Foonix", the above is very wrong. This is more correct (though still
2848 not perfect, because the below is a compile-time check):
2854 How does the HAS_QUUX become defined where it needs to be? Well, if
2855 Foonix happens to be UNIXy enough to be able to run the Configure
2856 script, and Configure has been taught about detecting and testing
2857 quux(), the HAS_QUUX will be correctly defined. In other platforms,
2858 the corresponding configuration step will hopefully do the same.
2860 In a pinch, if you cannot wait for Configure to be educated,
2861 or if you have a good hunch of where quux() might be available,
2862 you can temporarily try the following:
2864 #if (defined(__FOONIX__) || defined(__BARNIX__))
2874 But in any case, try to keep the features and operating systems separate.
2878 =head2 Problematic System Interfaces
2884 malloc(0), realloc(0), calloc(0, 0) are non-portable. To be portable
2885 allocate at least one byte. (In general you should rarely need to
2886 work at this low level, but instead use the various malloc wrappers.)
2890 snprintf() - the return type is unportable. Use my_snprintf() instead.
2894 =head2 Security problems
2896 Last but not least, here are various tips for safer coding.
2904 Or we will publicly ridicule you. Seriously.
2908 Do not use strcpy() or strcat() or strncpy() or strncat()
2910 Use my_strlcpy() and my_strlcat() instead: they either use the native
2911 implementation, or Perl's own implementation (borrowed from the public
2912 domain implementation of INN).
2916 Do not use sprintf() or vsprintf()
2918 If you really want just plain byte strings, use my_snprintf()
2919 and my_vsnprintf() instead, which will try to use snprintf() and
2920 vsnprintf() if those safer APIs are available. If you want something
2921 fancier than a plain byte string, use SVs and Perl_sv_catpvf().
2925 =head1 EXTERNAL TOOLS FOR DEBUGGING PERL
2927 Sometimes it helps to use external tools while debugging and
2928 testing Perl. This section tries to guide you through using
2929 some common testing and debugging tools with Perl. This is
2930 meant as a guide to interfacing these tools with Perl, not
2931 as any kind of guide to the use of the tools themselves.
2933 B<NOTE 1>: Running under memory debuggers such as Purify, valgrind, or
2934 Third Degree greatly slows down the execution: seconds become minutes,
2935 minutes become hours. For example as of Perl 5.8.1, the
2936 ext/Encode/t/Unicode.t takes extraordinarily long to complete under
2937 e.g. Purify, Third Degree, and valgrind. Under valgrind it takes more
2938 than six hours, even on a snappy computer-- the said test must be
2939 doing something that is quite unfriendly for memory debuggers. If you
2940 don't feel like waiting, that you can simply kill away the perl
2943 B<NOTE 2>: To minimize the number of memory leak false alarms (see
2944 L</PERL_DESTRUCT_LEVEL> for more information), you have to have
2945 environment variable PERL_DESTRUCT_LEVEL set to 2. The F<TEST>
2946 and harness scripts do that automatically. But if you are running
2947 some of the tests manually-- for csh-like shells:
2949 setenv PERL_DESTRUCT_LEVEL 2
2951 and for Bourne-type shells:
2953 PERL_DESTRUCT_LEVEL=2
2954 export PERL_DESTRUCT_LEVEL
2956 or in UNIXy environments you can also use the C<env> command:
2958 env PERL_DESTRUCT_LEVEL=2 valgrind ./perl -Ilib ...
2960 B<NOTE 3>: There are known memory leaks when there are compile-time
2961 errors within eval or require, seeing C<S_doeval> in the call stack
2962 is a good sign of these. Fixing these leaks is non-trivial,
2963 unfortunately, but they must be fixed eventually.
2965 B<NOTE 4>: L<DynaLoader> will not clean up after itself completely
2966 unless Perl is built with the Configure option
2967 C<-Accflags=-DDL_UNLOAD_ALL_AT_EXIT>.
2969 =head2 Rational Software's Purify
2971 Purify is a commercial tool that is helpful in identifying
2972 memory overruns, wild pointers, memory leaks and other such
2973 badness. Perl must be compiled in a specific way for
2974 optimal testing with Purify. Purify is available under
2975 Windows NT, Solaris, HP-UX, SGI, and Siemens Unix.
2977 =head2 Purify on Unix
2979 On Unix, Purify creates a new Perl binary. To get the most
2980 benefit out of Purify, you should create the perl to Purify
2983 sh Configure -Accflags=-DPURIFY -Doptimize='-g' \
2984 -Uusemymalloc -Dusemultiplicity
2986 where these arguments mean:
2990 =item -Accflags=-DPURIFY
2992 Disables Perl's arena memory allocation functions, as well as
2993 forcing use of memory allocation functions derived from the
2996 =item -Doptimize='-g'
2998 Adds debugging information so that you see the exact source
2999 statements where the problem occurs. Without this flag, all
3000 you will see is the source filename of where the error occurred.
3004 Disable Perl's malloc so that Purify can more closely monitor
3005 allocations and leaks. Using Perl's malloc will make Purify
3006 report most leaks in the "potential" leaks category.
3008 =item -Dusemultiplicity
3010 Enabling the multiplicity option allows perl to clean up
3011 thoroughly when the interpreter shuts down, which reduces the
3012 number of bogus leak reports from Purify.
3016 Once you've compiled a perl suitable for Purify'ing, then you
3021 which creates a binary named 'pureperl' that has been Purify'ed.
3022 This binary is used in place of the standard 'perl' binary
3023 when you want to debug Perl memory problems.
3025 As an example, to show any memory leaks produced during the
3026 standard Perl testset you would create and run the Purify'ed
3031 ../pureperl -I../lib harness
3033 which would run Perl on test.pl and report any memory problems.
3035 Purify outputs messages in "Viewer" windows by default. If
3036 you don't have a windowing environment or if you simply
3037 want the Purify output to unobtrusively go to a log file
3038 instead of to the interactive window, use these following
3039 options to output to the log file "perl.log":
3041 setenv PURIFYOPTIONS "-chain-length=25 -windows=no \
3042 -log-file=perl.log -append-logfile=yes"
3044 If you plan to use the "Viewer" windows, then you only need this option:
3046 setenv PURIFYOPTIONS "-chain-length=25"
3048 In Bourne-type shells:
3051 export PURIFYOPTIONS
3053 or if you have the "env" utility:
3055 env PURIFYOPTIONS="..." ../pureperl ...
3059 Purify on Windows NT instruments the Perl binary 'perl.exe'
3060 on the fly. There are several options in the makefile you
3061 should change to get the most use out of Purify:
3067 You should add -DPURIFY to the DEFINES line so the DEFINES
3068 line looks something like:
3070 DEFINES = -DWIN32 -D_CONSOLE -DNO_STRICT $(CRYPT_FLAG) -DPURIFY=1
3072 to disable Perl's arena memory allocation functions, as
3073 well as to force use of memory allocation functions derived
3074 from the system malloc.
3076 =item USE_MULTI = define
3078 Enabling the multiplicity option allows perl to clean up
3079 thoroughly when the interpreter shuts down, which reduces the
3080 number of bogus leak reports from Purify.
3082 =item #PERL_MALLOC = define
3084 Disable Perl's malloc so that Purify can more closely monitor
3085 allocations and leaks. Using Perl's malloc will make Purify
3086 report most leaks in the "potential" leaks category.
3090 Adds debugging information so that you see the exact source
3091 statements where the problem occurs. Without this flag, all
3092 you will see is the source filename of where the error occurred.
3096 As an example, to show any memory leaks produced during the
3097 standard Perl testset you would create and run Purify as:
3102 purify ../perl -I../lib harness
3104 which would instrument Perl in memory, run Perl on test.pl,
3105 then finally report any memory problems.
3109 The excellent valgrind tool can be used to find out both memory leaks
3110 and illegal memory accesses. As of August 2003 it unfortunately works
3111 only on x86 (ELF) Linux. The special "test.valgrind" target can be used
3112 to run the tests under valgrind. Found errors and memory leaks are
3113 logged in files named F<testfile.valgrind>.
3115 Valgrind also provides a cachegrind tool, invoked on perl as:
3117 VG_OPTS=--tool=cachegrind make test.valgrind
3119 As system libraries (most notably glibc) are also triggering errors,
3120 valgrind allows to suppress such errors using suppression files. The
3121 default suppression file that comes with valgrind already catches a lot
3122 of them. Some additional suppressions are defined in F<t/perl.supp>.
3124 To get valgrind and for more information see
3126 http://developer.kde.org/~sewardj/
3128 =head2 Compaq's/Digital's/HP's Third Degree
3130 Third Degree is a tool for memory leak detection and memory access checks.
3131 It is one of the many tools in the ATOM toolkit. The toolkit is only
3132 available on Tru64 (formerly known as Digital UNIX formerly known as
3135 When building Perl, you must first run Configure with -Doptimize=-g
3136 and -Uusemymalloc flags, after that you can use the make targets
3137 "perl.third" and "test.third". (What is required is that Perl must be
3138 compiled using the C<-g> flag, you may need to re-Configure.)
3140 The short story is that with "atom" you can instrument the Perl
3141 executable to create a new executable called F<perl.third>. When the
3142 instrumented executable is run, it creates a log of dubious memory
3143 traffic in file called F<perl.3log>. See the manual pages of atom and
3144 third for more information. The most extensive Third Degree
3145 documentation is available in the Compaq "Tru64 UNIX Programmer's
3146 Guide", chapter "Debugging Programs with Third Degree".
3148 The "test.third" leaves a lot of files named F<foo_bar.3log> in the t/
3149 subdirectory. There is a problem with these files: Third Degree is so
3150 effective that it finds problems also in the system libraries.
3151 Therefore you should used the Porting/thirdclean script to cleanup
3152 the F<*.3log> files.
3154 There are also leaks that for given certain definition of a leak,
3155 aren't. See L</PERL_DESTRUCT_LEVEL> for more information.
3157 =head2 PERL_DESTRUCT_LEVEL
3159 If you want to run any of the tests yourself manually using e.g.
3160 valgrind, or the pureperl or perl.third executables, please note that
3161 by default perl B<does not> explicitly cleanup all the memory it has
3162 allocated (such as global memory arenas) but instead lets the exit()
3163 of the whole program "take care" of such allocations, also known as
3164 "global destruction of objects".
3166 There is a way to tell perl to do complete cleanup: set the
3167 environment variable PERL_DESTRUCT_LEVEL to a non-zero value.
3168 The t/TEST wrapper does set this to 2, and this is what you
3169 need to do too, if you don't want to see the "global leaks":
3170 For example, for "third-degreed" Perl:
3172 env PERL_DESTRUCT_LEVEL=2 ./perl.third -Ilib t/foo/bar.t
3174 (Note: the mod_perl apache module uses also this environment variable
3175 for its own purposes and extended its semantics. Refer to the mod_perl
3176 documentation for more information. Also, spawned threads do the
3177 equivalent of setting this variable to the value 1.)
3179 If, at the end of a run you get the message I<N scalars leaked>, you can
3180 recompile with C<-DDEBUG_LEAKING_SCALARS>, which will cause the addresses
3181 of all those leaked SVs to be dumped along with details as to where each
3182 SV was originally allocated. This information is also displayed by
3183 Devel::Peek. Note that the extra details recorded with each SV increases
3184 memory usage, so it shouldn't be used in production environments. It also
3185 converts C<new_SV()> from a macro into a real function, so you can use
3186 your favourite debugger to discover where those pesky SVs were allocated.
3190 If compiled with C<-DPERL_MEM_LOG>, all Newx() and Renew() allocations
3191 and Safefree() in the Perl core go through logging functions, which is
3192 handy for breakpoint setting. If also compiled with C<-DPERL_MEM_LOG_STDERR>,
3193 the allocations and frees are logged to STDERR (or more precisely, to the
3194 file descriptor 2) in these logging functions, with the calling source code
3195 file and line number (and C function name, if supported by the C compiler).
3197 This logging is somewhat similar to C<-Dm> but independent of C<-DDEBUGGING>,
3198 and at a higher level (the C<-Dm> is directly at the point of C<malloc()>,
3199 while the C<PERL_MEM_LOG> is at the level of C<New()>).
3203 Depending on your platform there are various of profiling Perl.
3205 There are two commonly used techniques of profiling executables:
3206 I<statistical time-sampling> and I<basic-block counting>.
3208 The first method takes periodically samples of the CPU program
3209 counter, and since the program counter can be correlated with the code
3210 generated for functions, we get a statistical view of in which
3211 functions the program is spending its time. The caveats are that very
3212 small/fast functions have lower probability of showing up in the
3213 profile, and that periodically interrupting the program (this is
3214 usually done rather frequently, in the scale of milliseconds) imposes
3215 an additional overhead that may skew the results. The first problem
3216 can be alleviated by running the code for longer (in general this is a
3217 good idea for profiling), the second problem is usually kept in guard
3218 by the profiling tools themselves.
3220 The second method divides up the generated code into I<basic blocks>.
3221 Basic blocks are sections of code that are entered only in the
3222 beginning and exited only at the end. For example, a conditional jump
3223 starts a basic block. Basic block profiling usually works by
3224 I<instrumenting> the code by adding I<enter basic block #nnnn>
3225 book-keeping code to the generated code. During the execution of the
3226 code the basic block counters are then updated appropriately. The
3227 caveat is that the added extra code can skew the results: again, the
3228 profiling tools usually try to factor their own effects out of the
3231 =head2 Gprof Profiling
3233 gprof is a profiling tool available in many UNIX platforms,
3234 it uses F<statistical time-sampling>.
3236 You can build a profiled version of perl called "perl.gprof" by
3237 invoking the make target "perl.gprof" (What is required is that Perl
3238 must be compiled using the C<-pg> flag, you may need to re-Configure).
3239 Running the profiled version of Perl will create an output file called
3240 F<gmon.out> is created which contains the profiling data collected
3241 during the execution.
3243 The gprof tool can then display the collected data in various ways.
3244 Usually gprof understands the following options:
3250 Suppress statically defined functions from the profile.
3254 Suppress the verbose descriptions in the profile.
3258 Exclude the given routine and its descendants from the profile.
3262 Display only the given routine and its descendants in the profile.
3266 Generate a summary file called F<gmon.sum> which then may be given
3267 to subsequent gprof runs to accumulate data over several runs.
3271 Display routines that have zero usage.
3275 For more detailed explanation of the available commands and output
3276 formats, see your own local documentation of gprof.
3280 $ sh Configure -des -Dusedevel -Doptimize='-g' -Accflags='-pg' -Aldflags='-pg' && make
3281 $ ./perl someprog # creates gmon.out in current directory
3285 =head2 GCC gcov Profiling
3287 Starting from GCC 3.0 I<basic block profiling> is officially available
3290 You can build a profiled version of perl called F<perl.gcov> by
3291 invoking the make target "perl.gcov" (what is required that Perl must
3292 be compiled using gcc with the flags C<-fprofile-arcs
3293 -ftest-coverage>, you may need to re-Configure).
3295 Running the profiled version of Perl will cause profile output to be
3296 generated. For each source file an accompanying ".da" file will be
3299 To display the results you use the "gcov" utility (which should
3300 be installed if you have gcc 3.0 or newer installed). F<gcov> is
3301 run on source code files, like this
3305 which will cause F<sv.c.gcov> to be created. The F<.gcov> files
3306 contain the source code annotated with relative frequencies of
3307 execution indicated by "#" markers.
3309 Useful options of F<gcov> include C<-b> which will summarise the
3310 basic block, branch, and function call coverage, and C<-c> which
3311 instead of relative frequencies will use the actual counts. For
3312 more information on the use of F<gcov> and basic block profiling
3313 with gcc, see the latest GNU CC manual, as of GCC 3.0 see
3315 http://gcc.gnu.org/onlinedocs/gcc-3.0/gcc.html
3317 and its section titled "8. gcov: a Test Coverage Program"
3319 http://gcc.gnu.org/onlinedocs/gcc-3.0/gcc_8.html#SEC132
3323 $ sh Configure -des -Doptimize='-g' -Accflags='-fprofile-arcs -ftest-coverage' \
3324 -Aldflags='-fprofile-arcs -ftest-coverage' && make perl.gcov
3325 $ rm -f regexec.c.gcov regexec.gcda
3328 $ view regexec.c.gcov
3330 =head2 Pixie Profiling
3332 Pixie is a profiling tool available on IRIX and Tru64 (aka Digital
3333 UNIX aka DEC OSF/1) platforms. Pixie does its profiling using
3334 I<basic-block counting>.
3336 You can build a profiled version of perl called F<perl.pixie> by
3337 invoking the make target "perl.pixie" (what is required is that Perl
3338 must be compiled using the C<-g> flag, you may need to re-Configure).
3340 In Tru64 a file called F<perl.Addrs> will also be silently created,
3341 this file contains the addresses of the basic blocks. Running the
3342 profiled version of Perl will create a new file called "perl.Counts"
3343 which contains the counts for the basic block for that particular
3346 To display the results you use the F<prof> utility. The exact
3347 incantation depends on your operating system, "prof perl.Counts" in
3348 IRIX, and "prof -pixie -all -L. perl" in Tru64.
3350 In IRIX the following prof options are available:
3356 Reports the most heavily used lines in descending order of use.
3357 Useful for finding the hotspot lines.
3361 Groups lines by procedure, with procedures sorted in descending order of use.
3362 Within a procedure, lines are listed in source order.
3363 Useful for finding the hotspots of procedures.
3367 In Tru64 the following options are available:
3373 Procedures sorted in descending order by the number of cycles executed
3374 in each procedure. Useful for finding the hotspot procedures.
3375 (This is the default option.)
3379 Lines sorted in descending order by the number of cycles executed in
3380 each line. Useful for finding the hotspot lines.
3382 =item -i[nvocations]
3384 The called procedures are sorted in descending order by number of calls
3385 made to the procedures. Useful for finding the most used procedures.
3389 Grouped by procedure, sorted by cycles executed per procedure.
3390 Useful for finding the hotspots of procedures.
3394 The compiler emitted code for these lines, but the code was unexecuted.
3398 Unexecuted procedures.
3402 For further information, see your system's manual pages for pixie and prof.
3404 =head2 Miscellaneous tricks
3410 Those debugging perl with the DDD frontend over gdb may find the
3413 You can extend the data conversion shortcuts menu, so for example you
3414 can display an SV's IV value with one click, without doing any typing.
3415 To do that simply edit ~/.ddd/init file and add after:
3417 ! Display shortcuts.
3418 Ddd*gdbDisplayShortcuts: \
3419 /t () // Convert to Bin\n\
3420 /d () // Convert to Dec\n\
3421 /x () // Convert to Hex\n\
3422 /o () // Convert to Oct(\n\
3424 the following two lines:
3426 ((XPV*) (())->sv_any )->xpv_pv // 2pvx\n\
3427 ((XPVIV*) (())->sv_any )->xiv_iv // 2ivx
3429 so now you can do ivx and pvx lookups or you can plug there the
3430 sv_peek "conversion":
3432 Perl_sv_peek(my_perl, (SV*)()) // sv_peek
3434 (The my_perl is for threaded builds.)
3435 Just remember that every line, but the last one, should end with \n\
3437 Alternatively edit the init file interactively via:
3438 3rd mouse button -> New Display -> Edit Menu
3440 Note: you can define up to 20 conversion shortcuts in the gdb
3445 If you see in a debugger a memory area mysteriously full of 0xABABABAB
3446 or 0xEFEFEFEF, you may be seeing the effect of the Poison() macros,
3451 Under ithreads the optree is read only. If you want to enforce this, to check
3452 for write accesses from buggy code, compile with C<-DPL_OP_SLAB_ALLOC> to
3453 enable the OP slab allocator and C<-DPERL_DEBUG_READONLY_OPS> to enable code
3454 that allocates op memory via C<mmap>, and sets it read-only at run time.
3455 Any write access to an op results in a C<SIGBUS> and abort.
3457 This code is intended for development only, and may not be portable even to
3458 all Unix variants. Also, it is an 80% solution, in that it isn't able to make
3459 all ops read only. Specifically it
3465 Only sets read-only on all slabs of ops at C<CHECK> time, hence ops allocated
3466 later via C<require> or C<eval> will be re-write
3470 Turns an entire slab of ops read-write if the refcount of any op in the slab
3471 needs to be decreased.
3475 Turns an entire slab of ops read-write if any op from the slab is freed.
3479 It's not possible to turn the slabs to read-only after an action requiring
3480 read-write access, as either can happen during op tree building time, so
3481 there may still be legitimate write access.
3483 However, as an 80% solution it is still effective, as currently it catches
3484 a write access during the generation of F<Config.pm>, which means that we
3485 can't yet build F<perl> with this enabled.
3492 We've had a brief look around the Perl source, how to maintain quality
3493 of the source code, an overview of the stages F<perl> goes through
3494 when it's running your code, how to use debuggers to poke at the Perl
3495 guts, and finally how to analyse the execution of Perl. We took a very
3496 simple problem and demonstrated how to solve it fully - with
3497 documentation, regression tests, and finally a patch for submission to
3498 p5p. Finally, we talked about how to use external tools to debug and
3501 I'd now suggest you read over those references again, and then, as soon
3502 as possible, get your hands dirty. The best way to learn is by doing,
3509 Subscribe to perl5-porters, follow the patches and try and understand
3510 them; don't be afraid to ask if there's a portion you're not clear on -
3511 who knows, you may unearth a bug in the patch...
3515 Keep up to date with the bleeding edge Perl distributions and get
3516 familiar with the changes. Try and get an idea of what areas people are
3517 working on and the changes they're making.
3521 Do read the README associated with your operating system, e.g. README.aix
3522 on the IBM AIX OS. Don't hesitate to supply patches to that README if
3523 you find anything missing or changed over a new OS release.
3527 Find an area of Perl that seems interesting to you, and see if you can
3528 work out how it works. Scan through the source, and step over it in the
3529 debugger. Play, poke, investigate, fiddle! You'll probably get to
3530 understand not just your chosen area but a much wider range of F<perl>'s
3531 activity as well, and probably sooner than you'd think.
3537 =item I<The Road goes ever on and on, down from the door where it began.>
3541 If you can do these things, you've started on the long road to Perl porting.
3542 Thanks for wanting to help make Perl better - and happy hacking!
3546 This document was written by Nathan Torkington, and is maintained by
3547 the perl5-porters mailing list.